Random crashes while gaming

Hey all, I need some help with figuring out what's crashing my PC, because I've tried everything I know and searched a lot on the internet, but haven't been able to solve it.
My PC randomly crashes when playing various games. This has been going on for about a year. The crashes are always the same: PC shuts down completely with no warning and does not reboot on its own. I need to press the power button, after which it immediately starts as if nothing ever happened. This does not happen with every game and it also does not necessarily happen when playing very demanding games. I will give a brief overview of some of the games that caused problems and how I dealt with it.

- Frostpunk: crashes every 30 minutes, or sometimes within a few minutes. Was able to fix this by turning all settings to low. No crashes anymore.
- Farthest Frontier: crashes every few minutes. Fixed it by setting my GPU max clock speed to -20% in my drivers.
- Black Myth: Wukong: First played it for about an hour and a half with no problems. Today I played it for half an hour after which it crashed. I rebooted and started the game up again and it crashed within a few minutes.

I haven't tried it in a while, but when this problem first occurred I was also playing Elden Ring (which is much more demanding than Frostpunk and Farthest Frontier) and did not encounter any problems with that. Currently the only game I play a lot is Rocket League, which has never had any problems.

To further understand the problem, I did some stress tests:
- in-browser stress test: gpu, cpu and combined. All 3 for 5-10 minutes, no crash.
- FurMark 2 stress test:

• crash after 2-3 min
• set max clock speed to -20%. No crash after 9 minutes of testing
• set max clock back to 0. After 4 minutes of no crash, I added the in-browsed cpu stress test. Crash after 10 sec. The fact that adding the cpu caused it to crash, made me think that the problem is with the PSU rather than the GPU, but a brand new PSU was not able to fix the issue.
• I noticed that when I start a stress test with a cool PC, the GPU wattage fluctuates heavily (average 120, lows of 100, frequent peaks of 170-200). However, after about 5 minutes, it stabilises to about 120, with very few fluctuations, possibly due to thermal throttle (though it reaches max temp almost immediately). When the clock speed is set to -20%, the wattage is stable at 120 from the get go.

Things I've tried: brand new PSU, including new cables to everything except peripherals. Updated drivers.

Specs:
CPU: AMD Ryzed 5 5600X
GPU: Gigabyte Radeon AMD RX 5700 XT Gaming OC 8G
RAM: Corsair Vengeance LPX DDR4 3200 MHz, 2x8GB
MOBO: Gigabyte x470 Aorus ultra gaming
PSU: Corsair RM 850x (850 watt)

Any help in figuring out what's going on would be greatly appreciated!

Thanks for clarifying the difference between crashes and shutdowns and thanks for your help in trying to solve the issue.
I have set the Windows Power Plan to Balanced (it was at high performance) and I have restored the BIOS settings to default settings.
After about 10 minutes of playing Black Myth: Wukong, it shut down in the same manner as usual.

The PC is, and always has been, plugged into a power strip.
I have run the MiniToolBox program according to your instructions and have attached the output.
In terms of overheating, I can already tell you that since I've had this GPU the hotspot temperature has always immediately shot up to the maximum 110 degrees Celsius under load. For the first two years or so, however, I had no shutdowns, despite the temperature being 110 for hours on end.

I used the AMD drivers to check the temperature.
Currently using non legit Windows 10. Used to be legit but an update gave me problems and after rolling it back it didn't accept my license. Decided pirating was the easiest fix.