[clang] [Sema] Mark alias/ifunc targets used and consider mangled names (PR #87130)

[Nick Desaulniers via cfe-commits]

https://github.com/nickdesaulniers commented:

As the author of https://reviews.llvm.org/D54188 and the TODO added from there, this patch is the right thing to do.  I fixed a false positive diagnostic for C and knew that the fix was insufficient for C++.

I do appreciate the abundance of caution here, but this looks generally like what we want to do.

>  We fix false positives while causing false negatives when overloads are present.

Can we avoid the false negatives?

https://github.com/llvm/llvm-project/pull/87130