[llvm] [AMDGPU] Eliminate unnecessary packing in wider f16 vectors for sdwa/opsel-able instruction (PR #137137)

[Pierre van Houtryve via llvm-commits]

https://github.com/Pierre-vh commented:

I'm not familiar enough with the changes here so I can't review the feature itself. I limited my review to general coding style remarks

https://github.com/llvm/llvm-project/pull/137137