[llvm] [AMDGPU] Handle unset/max flat workgroup size in waves/EU (PR #139955)

[Shilei Tian via llvm-commits]

https://github.com/shiltian requested changes to this pull request.

I'm gonna put a block here, since this block of code is gonna go anyway after https://github.com/llvm/llvm-project/pull/123995 is merged.

https://github.com/llvm/llvm-project/pull/139955