Ouch you know what not being funny here just a thought since you say it has done it since day one of the build. I would have a close check of the gold pins on the under side of the Cpu pw2374, it looks like a pin maybe slightly bent, or worse broke off ect. By most of the reports it has something to do with the CPU.
The other exception to do with the ATI card drivers Has to do with an Irq conflict.
Each device is assigned its own IRQ number to talk directly to the Cpu. if it failing it can be a conflict with other hardware, or sign the Cpu is at fault. All and everything has to go through the Cpu in the list so if you think a bit logic here check the Cpu out for bent pins ect then do a re-seat of it see if it helps. You may want to look into the PSU aswell it can also be the cause of things you are listing if it is fluctuating in power and not steady.It can cause problems with ram ect when fluctuating causing errors. You never listed its brand and wattage. so if you can do so it may help so we can see if it is the culprit also pw2374. That is my two cents worth if it helps you in any way with some insight to the problem.
amdppm.sys: seems to be a stack over write/ minidump issue. It could be malware causing a minidump as part of an exploit. Might want to run Driver Verifier to see what the actual bad boy is for this. I have also see Avast cause this. This is probably what is causing your memory BSODs
Memory is my bet. I'd try memtest86+ and if that doesn't find problems, start swapping out modules while trying to replicate the crash.
There are way too many different BSODs here for it to be related to pretty much anything else. There are also too many "causes" for them to make it seem plausible that those drivers are actually at fault. As Groundrat suggested, you could try Driver Verifier but I believe it's not going to be of much value if the "bugs" it finds are due to memory corruption.
DRIVER_IRQL_NOT_LESS_OR_EQUAL / IRQL_NOT_LESS_OR_EQUAL - Accessing paged memory from above APC_LEVEL. Could happen if the pointer is stored in a faulty memory area.
SYSTEM_THREAD_EXCEPTION_NOT_HANDLED / KMODE_EXCEPTION_NOT_HANDLED / SYSTEM_THREAD_EXCEPTION_NOT_HANDLED - Unhandled exceptions. Probably generated from code regions that the developers did not expect to throw them. Memory corruption is a likely reason for this too.
MEMORY_MANAGEMENT / CACHE_MANAGER - Both indicate critical failures of memory management. Cache manager integrates tightly with memory manager, so I'm not surprised it failed as well.
PAGE_FAULT_IN_NONPAGED_AREA - Access to invalid memory again. Also likely result of memory corruption.
NTFS_FILE_SYSTEM - Not quite sure on this one but it's plausible that it suffered an internal failure due to memory corruption.