[llvm] [AMDGPU] Extend wave reduce intrinsics for i32 type (PR #126469)

[Joseph Huber via llvm-commits]

https://github.com/jhuber6 commented:

Do these implement the prefix sum / scan? Given that the last value of the scan is the reduction I'm assuming that's the case.

https://github.com/llvm/llvm-project/pull/126469