[pinephone v1.2/kernel bug] Thermal throttling appears to be disabled entirely, should be enabled

I have noticed the thermal throttling kicks in too softly under full, longer cpu load.

Steps to reproduce:

1. run any application that actually makes use of all CPU cores, e.g. desktop firefox playing a youtube video with likely software decoding on the youtube desktop site with its complex layout. doesn't hurt if it also does some GPU usage at the same time which I'm sure desktop firefox also does
2. at the same time, charge the phone with a 5V 2A charger
3. make sure to do this for a while, minimum of ten minutes non-stop. You will think at first it's pretty hot but in an understandable range, and it might feel like it plateaus before it gets critical. However, you will realize further in that you thought it had stopped getting even insanely hotter when in fact it hasn't, and it's sneaking up to even further unholy levels

Expected result: significant throttling once temperatures get really hot

Actual result: no significant or insufficient throttling, and device reaches unacceptable temperature ranges. The result of this thermal fallout is so bad, I have among other things seen in separate instances:

• hard emergency shutoff where the whole CPU just turns off without warning (I assume the 70 degree hard limit or whatever it is, or even a CPU crash)
• significant screen discoloration due to the heat behind it with pixels inverting in colors and other insanity, which thankfully appeared to be temporary and went away when cooling it down again
• phone backside so hot I have slightly burned my hand (no permanent blisters, but hurt for quite a while after even though I touched it as little as possible to get it outside the cold window to cool down)

This is not only a human safety problem due to how hot it gets to touch, but also risking to greatly degrade battery and CPU life. Therefore, I think this really needs addressing. From asking around in chat and having others test this scenario briefly, I believe this to likely not be a hardware defect. Rather, many users don't seem to have run into this yet because they don't run software that puts full load on all cores for 10+ minutes nonstop.

Please note I don't care about the short term battery life here (like what crust would solve) since obviously that's always going to tank under full load. This is purely about the overheating issues.

Where should I make an upstream ticket for this? (Since I assume this is more of a kernel-wide rather than a postmarketOS-specific problem)