Although this is true for most games it is worth noting that it isn't universally true. Usermode anti-cheat does sometimes work verbatim in Wine, and some anti-cheat software has Proton support, though not all developers elect to enable it.
You can be clever and build a random memory allocator. You can get clever and watch for frozen struct members after a known set operation, what you can’t do is prevent all cheating. There’s device layer, driver layer, MITM, emulation, and even now AI mouse control.
The only thing you can do is watch for it and send the ban hammer. Valve has a wonderful write up about client-side prediction recording so as to verify killcam shots were indeed, kill shots, and not aim bots (but this method is great for seeing those in action as well!)
Sure, but you still have to make a serious attempt or the experience will be terrible for any non-cheaters. Or you just make your game bad enough that no one cares. That's an option too.
If you don’t need real-time packets and can deal with the old school architecture of pulses, there’s things you can do on the network to ensure security.
You do this too on real-time UDP it’s just a bit trickier. Prediction and analysis pattern discovery is really the only options thus far.
But I could be blowing smoke and know nothing about the layers of kernel integration these malware have developed.
Kernel level? The SOTA cheats use custom hardware that uses DMA to spy on the game state. There are now also purely external cheating devices that use video capture and mouse emulation to fully simulate a human.