[llvm] [AMDGPU] Introduce "amdgpu-uniform-intrinsic-combine" pass to combine uniform AMDGPU lane Intrinsics. (PR #116953)

[Sameer Sahasrabuddhe via llvm-commits]

https://github.com/ssahasra approved this pull request.

I do believe that we had already agreed to make it a ModulePass because traversing over the uses of the intrinsic declaration is cheaper than traversing over every function body looking for calls to very specific intrinsics.

LGTM, please wait 24 hours before submitting so that other reviewers have a chance to comment.

https://github.com/llvm/llvm-project/pull/116953