[clang] [Clang][OpenCL][AMDGPU] Allow a kernel to call another kernel (PR #115821)

[John McCall via cfe-commits]

rjmccall wrote:

It's not actually clang's responsibility to not break your transformation. We don't want to break it gratuitously, but if clang is generating correct code, your transformation needs to be flexible enough to handle it. If there's something easy we can do to help your transformation continue to work while still compiling the source language properly, of course, we're absolutely open to requests.

https://github.com/llvm/llvm-project/pull/115821