[llvm] [mlir] [llvm][OpenMP] Handle complex types in atomic read (PR #111377)

[via llvm-commits]

NimishMishra wrote:

Hi @tblah,

   I have tested out this PR with working on https://github.com/llvm/llvm-project/issues/112908, and I do not see segfault (once we remove `do`). Is `atomic read` still segfaulting on aarch64?

https://github.com/llvm/llvm-project/pull/111377