@TobyXR6 This is sounding incredibly similar. I've traced mine back to the same thing - a lack of grounding on pin 926. However I've never had the starter motor fail to activate (yet...). I also haven't noticed any correlation with heat. More often than not mine occurs on a cold start when the car has been sitting for a while (not daily driven).
I was planning to do the same and install a manual override switch for the fuel pump but was also concerned about safety. My idea was using a two stage switch, where you had a toggle switch run in series with a signal that was only live when the engine was running (first thing that comes to mind is oil pressure switch), and then a push button switch that would be a full override? Flip the switch, hold the push button to start (100% override), and then the toggle switch + oil pressure switch will keep the fuel pump switched on as long as the engine is running. This would at least cut the pump in a scenario such a broken fuel line. You also have the inertia switch which would cut the pump in certain types of accidents, and that still functions as normal with the relay manually switched on.
As for the P0230 DTC being caused by your override switch, I think it's a definite possibility. Unless there is some other PCM connection to the secondary circuit that I'm not aware of, the PCM doesn't really know whether the fuel pump is on or not, but your override switch would be causing the PCM to "see" a ground that it didn't ask for. If you wanted to fix this, what if you added your own independent fuel pump circuit, by running your override wire to a new relay instead of the existing one and leaving the PCM out of it completely?
Hearing that your PCM has been bench tested as normal gives me a bit of hope to keep diagnosing this instead of replacing my PCM now.
@BarraT Could be on the right track with the grounding issue. @TobyXR6 Have you checked all your grounds yet? This is the next thing on my list. Maybe your issue is a bad ground somewhere that becomes more interrupted with thermal expansion around it? That would explain the heat correlation.
One interesting thing I observed was that if I kept the car on a trickle charger and kept the battery voltage maxed out, it seems to happen a lot less frequently. Have you looked into battery voltages / battery health / parasitic drain at all? Does the car hold charge long term or will it drain over time? I know mine has a small drain, narrowed down to something on the BEM fuse (haven't diagnosed exactly what yet).
I realise all these suggestions are band aid fixes and don't really work towards finding the actual issue, but hopefully some of them are still helpful.
