Adamw

With innovate controllers the sensor usually hasnt failed, they just refuse to work when the frequency response gets slower.  If you get error 2 or error 8 the sensor is likely still perfectly fine and will work with any other wideband controller, if you haven't thrown away already, save them as spares for when you have a decent controller.   
Innovate determine oxygen content by driving the sensor completely differently than they were ever designed to be, rather than measuring the actual pump current required to keep the nerst cell near stoich they instead drive the pump at full current all the time, continuously bouncing current backward and forwards, then use the resulting duty cycle to derive how rich or lean it was.  This can give very fast response and works fine when the sensor is new, clean, and its temperature is relatively stable, but age, temperature and deposits effect the timing a lot so the controller then justs reports the sensor has failed if it takes too long to see the response it expects. 
    

The log files doesnt seem to match the map.  For example the lambda target is 1.00 in the log but it is 0.905 in the map.  
The lambda is not even close to target so that would be the first thing to correct.  Why is the injector size set to 1600cc if it has 1700's?
Also, ID1700's dont have very good short PW behavior and for this reason ID dont recommend using them with gasoline where low flow is important, so dont expect miracles. 

It lives! (runs terribly of course, but it starts!)

I wired in a 2 channel igniter into my adapter loom and hey presto we have ignition. So embarrassing that I missed that... Thanks for the help team! 

I agree with Derek, just not enough air.  It appears the valve is working correctly as we see the position reduce as the engine warms up.  Setting the base position table as he suggests will work for warm but you are already needing 100% for anything less than 50°C so its not going to have enough air for idle target on cold start.  It might be better to adjust the throttle stop (assuming it is adjustable?) so the idle valve doesnt need to flow so much.  
 

The oscillation is closed loop lambda, mostly the update rate is too high, possibly excessive gain as well, but start with the rate.  Reduce the update rate to 1Hz at 1000RPM, 2Hz at 2000RPM and maybe 5Hz at 3000. 
 
 The overshoot is excessive base position and fan step I think.  If we look at time section 2:25, you are idling close to target RPM with 2 fans running with the throttle open 4.5%.  But you have 4% commanded in E-throttle target, plus 2.0% in base position, and a further 1.5% fan step, so you are telling the ecu to go to 7.5%, when it only needs 4.5%.   You can see the "E-throttle ISC CL trim" is pulling 3% out to achieve the idle target.   The oscillation is initially caused again by CLL, but this does then cause the RPM to drop low enough to hit the anti-stall gain which upsets things further.  Reduce the anti-stall gain to 1.5 or 2.0.  
 
This is the same CLL and anti-stall gain issue as above.  
 
Not really related to idle but something I noticed in your log - the engine fan 2 only has a 1deg hysteresis on it which causes the fan to bounce on and off rapidly when the temp crosses the threshold.  ECT only has a 1 deg resolution so your hysteresis needs to be bigger.  This will be pretty hard on the relay and fan motor.  

I wouldnt use the 0.001uF between pin B & C, but the other one on the main power supply is a nice precaution to take.  You just need one on the main power supply each side of the engine as close as possible to the coils.  Many dont bother and I rarely see any issue from them so dont sweat it too much, can always add one later if you have some noise issue.  

Yeah I dont really like the way the base position table has been set up in this, and it is in open loop too, so the ecu cant do anything to correct it.  I think this is likely most of your problem.
Attached is a copy of your map with some changes to your idle settings.  Can you load this in and do a similar log from cold start.  It will likely still need some adjustments but with closed loop working it should give us a better log to work from.  
R32 GTS4 MAP 403HP idle tweaks.pclx

Pin 16 is not controlled by the ecu, it is a simple FET circuit on the adapter board that just grounds pin 16 when pin 45 has ign signal on it.  This is why I want to know the voltages so we can determine which side the issue is coming from.  Most likely it is something that has been wired to the ign switch circuit holding pin 45 high.  
I cant imagine any scenario that the diode will solve.  1 side of the relay coil's should go straight to battery live.  The other side to pin 16.  For the relay to energise the ecu has to ground pin 16 so current flows from the battery, through the relay coil and back to ground.   A diode is basically a "one way valve", so the only way the diode could work was if the current was somehow reversed when you switched off ignition - that would mean the relay coil would have to be disconnected from the battery and connected to ground instead and the ecu would have to be sending voltage out of pin 16 (there is no internal path from pin 16 to battery).  The only explanation I can think of would be either the relays arent engaging at all now and instead everything is being powered up by a back feed through something else, or there is something else besides the ecu controlling the pin 16 wire, or those relays have been rewired so the 12+ supply side is not connected straight to battery anymore.  

Generally you dont want the ecu to control idle while the car is moving.  The problem with having the idle control activating at a higher speed is when you are coasting in overrun (in gear etc), the wheels are driving the engine, the ecu can not control the engine speed by closing the throttle.   So what happens is the ecu sees the RPM is above target so it closes the throttle more and more and retards the ignition trying to bring it down to target, but this has little effect since the car is driving the engine.  As soon as you push in the clutch the wheels are no longer holding the RPM up, and because the throttle is now way too closed the engine instantly stalls.  
It is not too common to shift to neutral during overrun so it is not something I have had to consider much.  I dont really see it being an issue if the idle was a little high when coasting in neutral?  
Its hard to tell much from your log without knowing what the driver was doing etc, but the common theme I see in it is you get a big undershoot when approaching idle then it bounces quite a bit higher than target.  I think the undershoot is probably coming from over-run fuel cut and retard.  Try turning that off or setting its deactivation say 500RPM higher so that is not interfering while you experiment with dashpot and idle approach.  
Note the 350Z does have the clutch and brake switch wired in series to DI6 (used for cruise cancel) so you may be able to use that to add some further logic.  Your neutral switch doesnt show any change of state in your log so im not sure it maybe it is only auto's that have that or if it isn't setup correctly.  
For the accel fuel I think it may need higher sensitivity in the cells 2000RPM and below, and probably a bit more cold correction in the 40-60°C area. 

Generally you dont want the ecu to control idle while the car is moving.  The problem with having the idle control activating at a higher speed is when you are coasting in overrun (in gear etc), the wheels are driving the engine, the ecu can not control the engine speed by closing the throttle.   So what happens is the ecu sees the RPM is above target so it closes the throttle more and more and retards the ignition trying to bring it down to target, but this has little effect since the car is driving the engine.  As soon as you push in the clutch the wheels are no longer holding the RPM up, and because the throttle is now way too closed the engine instantly stalls.  
It is not too common to shift to neutral during overrun so it is not something I have had to consider much.  I dont really see it being an issue if the idle was a little high when coasting in neutral?  
Its hard to tell much from your log without knowing what the driver was doing etc, but the common theme I see in it is you get a big undershoot when approaching idle then it bounces quite a bit higher than target.  I think the undershoot is probably coming from over-run fuel cut and retard.  Try turning that off or setting its deactivation say 500RPM higher so that is not interfering while you experiment with dashpot and idle approach.  
Note the 350Z does have the clutch and brake switch wired in series to DI6 (used for cruise cancel) so you may be able to use that to add some further logic.  Your neutral switch doesnt show any change of state in your log so im not sure it maybe it is only auto's that have that or if it isn't setup correctly.  
For the accel fuel I think it may need higher sensitivity in the cells 2000RPM and below, and probably a bit more cold correction in the 40-60°C area. 

What specific wideband controller do you have?

That would suggest you are using the wrong trigger mode, possibly the multitooth/missing mode?  The correct trigger mode for this engine is "1JZ VVTi" and the offset should be near zero.   You will have unreliable sync/no start/long startup and sometimes VVT errors if you are trying to use "cam pulse 1X" sync mode with a 3 tooth cam. 

Normal scheduled fuel injection is synchronous - that is it only gets injected once at a fixed point in the engine cycle (usually finishing around 400deg BTDC, so a long time before combustion).  So if you only have synchronous injection (like most ecus) and you suddenly stand on the throttle after the injection event has already finished, then the next combustion event will be lean because you only originally added enough fuel for the amount of air that was flowing at the time of injection.  
With Asynchrous injection the ecu is always watching the fuel calculation even after the injection event, so if there is a sudden change in airflow after the main injection has finished, then ecu will add extra injection events to "top up" the fuel that is already sitting in the port waiting to be inhaled.
The minimum async injection setting is only used to prevent the async from reactivating too often when there is only a small amount of extra fuel calculated.  I suspect the main thing that influences what is a good minimum setting is your injector size.  Probably around 1.0ms would be suitable for typical ~500cc injectors, and maybe 0.5ms for ~1000cc injectors.  All my cars started with 0.5ms in there and it seems to work good so I havent experimented with it much further.
Your little lean spike could be not ernough accel fuel or the min async is too big.  You really need to look at what was happening to accel fuel at the same time.  Also, dont be too concerned about getting a dead flat lambda, there are all sorts of interactions going on during heavy transients and measured lambda is not always a good indication of how well it is all working.  Aim for best response rather than flatest lambda.  
  
  
 

The VVTi camshaft has 3 teeth on it so multitooth/missing trigger with cam pulse 1X sync mode wont work unless you no longer have the VVT cams? 
Trigger mode should normally be 1JZ vvti mode for this engine.  
What do you have for coils/ignitors?

You need to use Subaru EJ20AVCS vvt mode.  An EJ207 has 4 teeth on the cam, an EJ25 has 3 teeth. 
Once that is changed then use the cam angle test on each side to set the correct offset on the DI's.  Instructions are in the help file page:  G4+ ECU Tuning Functions > VVT Control > Tuning VVT Control > Cam Angle Test

It will be disabled until engine is running and a short delay after startup. 

Files attached.  The fuel pump and 2 fans are now set up in the ECU as virtual aux 1-3, these are sent to the PDM in a user stream, all the usual channels are sent in the default MXS stream.  
The .xc1 file attached you can import into the CAN protocol builder in RS3 if you want to add anything new later.  
ZipTieTECH Zenki S14 V1.1.zconfig LINK_ZipTie Tech_@20230620_105433_005544.xc1 Unrivaled_Tuning V1.1.pclr

Your general ecu setup looks ok, I would suggest copying the charge temp approximation table from our V11 base map, but be aware the fuel map will need to be redone after that.  
Your fuel map doesnt look too bad to me, subaru's tend to have quite a bit of fuel pressure resonance in the rails that get worse with larger injectors and this often causes ridges in the fuel map.  
Also be aware the surface view scales the Z axis range to span the full height which tends to exaggerate a VE table that has a relatively small range of values.  The smallest value in your table is about 62 and the largest is about 110 so it is quite exaggerated by this effect.  If you temporarily enter 0 in one corner of your table and 150 in another corner you will see that the spikes and ridges dont look so bad.  
Original table left Vs the same table spanned full range on the right:
  

Your general ecu setup looks ok, I would suggest copying the charge temp approximation table from our V11 base map, but be aware the fuel map will need to be redone after that.  
Your fuel map doesnt look too bad to me, subaru's tend to have quite a bit of fuel pressure resonance in the rails that get worse with larger injectors and this often causes ridges in the fuel map.  
Also be aware the surface view scales the Z axis range to span the full height which tends to exaggerate a VE table that has a relatively small range of values.  The smallest value in your table is about 62 and the largest is about 110 so it is quite exaggerated by this effect.  If you temporarily enter 0 in one corner of your table and 150 in another corner you will see that the spikes and ridges dont look so bad.  
Original table left Vs the same table spanned full range on the right:
  

It still looks like just a calibration with your fuel pressure to me.  Regardless of which log I check and at what load I check your data returns a repeatable sensor slope of about 225kpa/V.  That doesnt quite fit the commonly available sensor ranges  - a 100psi sensor for example would be about 172kpa/V or a 150psi sensor would be ~259kpa/V.  It seems to be quite repeatable suggesting the mechanical system is working correctly. There would be some error due to regulator gradient and transient delays so I would say it is most likely a 150Psi or 1000Kpa calibration.  
Here is a snip of your log with a math channel used to generate differential FP if it had a 150psi sensor calibration set (Yellow trace).  It looks far more realistic and a FP of 167kpa would fit your symptoms.  

Turn on individual cyl correction.  You can either do a single value for each rotor or a table for each rotor.  

Ign 3 is ok, but Ign 4 is interconnected with another circuit that could make it behave erratically.  DI3 is also linked to the relay control circuit on Ign 4.  So for instance if you put 12V into DI3 or turn on its pull-up resistor you will find ign 4 wont work.

There is something up with your MAP sensor also, it takes 2 seconds to creep back to atmosphere after the engine has stopped. 

It is nothing to do with fuel or cold start enrichment.  You have a 100% ignition cut because you have enabled a MAP limit which is set lower than atmospheric pressure.