[clang] [Clang] Yet more consistent Unicode diagnostics (PR #211002)

[Corentin Jabot via cfe-commits]

================
@@ -133,20 +133,20 @@ def err_invalid_utf8 : Error<
 def warn_invalid_utf8_in_comment : Extension<
   "invalid UTF-8 in comment">, InGroup<DiagGroup<"invalid-utf8">>;
 def err_character_not_allowed : Error<
-  "unexpected character <U+%0>">;
+  "unexpected character %0">;
----------------
cor3ntin wrote:

Consistency, we decided to go with `'C' U+xxx` as syntax in another PR.
We still use a `<>` separator when we want to display a codepoint in a whole string or to display invalid bytes 

https://github.com/llvm/llvm-project/pull/211002