[llvm] [AMDGPU] Document & Finalize GFX12 Memory Model (PR #98599)

[Pierre van Houtryve via llvm-commits]

Pierre-vh wrote:

@nhaehnle @jayfoad I added a comment in the `insertRelease` about the approach we follow for WB emission.


https://github.com/llvm/llvm-project/pull/98599