[clang] [llvm] AMDGPU: Move enqueued block handling into clang (PR #128519)

[Sameer Sahasrabuddhe via cfe-commits]

================
@@ -1,23 +1,23 @@
-//===- AMDGPUOpenCLEnqueuedBlockLowering.h -----------------------*- C++-*-===//
+//===- AMDGPUExportKernelRuntimeHandles.h -----------------------*- C++-*-===//
----------------
ssahasra wrote:

Do we need the filename and the emacs marking on the first line anymore? I would be very happy to see it dropped.

https://github.com/llvm/llvm-project/pull/128519