[llvm] [mlir] [AMDGPU][MLIR]Add shmem-optimization as an op using transform dialect (PR #81550)

[via llvm-commits]

================
@@ -220,6 +220,21 @@ mlir::amdgpu::optimizeSharedMemoryReadsAndWrites(Operation *parentOp,
   return success();
 }
 
+void mlir::amdgpu::optimizeSharedMemoryReadsAndWritesOp(
+    ::mlir::func::FuncOp funcOp) {
----------------
harsh-nod wrote:

Do you need the ::mlir namespace?

https://github.com/llvm/llvm-project/pull/81550