[clang] [Clang] Add `__builtin_[ordered_]reduce_addf` for ordered/unordered fp reductions (PR #176160)

[Benjamin Maxwell via cfe-commits]

================
@@ -1136,6 +1136,7 @@ RValue CIRGenFunction::emitBuiltinExpr(const GlobalDecl &gd, unsigned builtinID,
   case Builtin::BI__builtin_reduce_xor:
   case Builtin::BI__builtin_reduce_or:
   case Builtin::BI__builtin_reduce_and:
+  case Builtin::BI__builtin_reduce_addf:
----------------
MacDue wrote:

I'll rename it to `fadd`. There does not really seem to be a consistent rule here. I was looking at C functions like `modf` (which overloads to `modf`, `modff` and `modfl`). 

https://github.com/llvm/llvm-project/pull/176160