[clang] [CIR] Fix GlobalOp::getSuccessorRegions using wrong region for dtor (PR #185115)

[Henrich Lauko via cfe-commits]

xlauko wrote:

> lgtm -- Is there any way we could add a test to catch this kind of thing?

Either as part of some (data/control flow) analysis, I don't think we have anything like that upstreamed.
But one can test this with DCE for example.

But we shoud also probably test Region interfaces thoroughly as mlir does: https://github.com/llvm/llvm-project/blob/main/mlir/lib/Interfaces/ControlFlowInterfaces.cpp 

https://github.com/llvm/llvm-project/pull/185115