[clang] [AMDGPU] Use generic builtins for `wave_reduce` ops (PR #179589)

[Matt Arsenault via cfe-commits]

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

These aren't "generic builtins".

Also the FP and integer operations should absolutely not share a function name. Especially min and max 

https://github.com/llvm/llvm-project/pull/179589