[llvm] [AMDGPU] Simplify and improve codegen for llvm.amdgcn.set.inactive (PR #107889)

[Jay Foad via llvm-commits]

jayfoad wrote:

> Resulting ISA is basically the same for GFX10+?

Yes. I tested GFX11 on graphics shaders and saw no differences at all. The original motivation for this was removing the complexity in `expandPostRAPseudo`. Any codegen improvements for GFX9 are just a bonus.

https://github.com/llvm/llvm-project/pull/107889