[llvm] [MemProf] Reduce cloning overhead by sharing nodes when possible (PR #99832)

[Snehasish Kumar via llvm-commits]

================
@@ -1757,6 +1815,7 @@ IndexCallsiteContextGraph::IndexCallsiteContextGraph(
           if (AN.MIBs.empty())
             continue;
           CallsWithMetadata.push_back({&AN});
+          CallToFunc[{&AN}] = FS;
----------------
snehasish wrote:

Sure, it is a little clearer. There are other places which should also be updated to make it consistent so I don't have a strong opinion.

https://github.com/llvm/llvm-project/pull/99832