[llvm] [memprof] Improve deserialization performance in V3 (PR #94787)

[Kazu Hirata via llvm-commits]

================
@@ -499,12 +500,56 @@ void CallStackRadixTreeBuilder::build(
 
   // Sort the list of call stacks in the dictionary order to maximize the length
----------------
kazutakahirata wrote:

Fixed in the latest iteration.  I've removed the old comment and rephrased the immediately following paragraph.

https://github.com/llvm/llvm-project/pull/94787