clarify comparison of output: ignore token positions, which are somewhat accidental;

--- a/src/Pure/Syntax/parser.ML	Mon Sep 30 11:42:52 2024 +0200
+++ b/src/Pure/Syntax/parser.ML	Mon Sep 30 12:59:50 2024 +0200
@@ -574,7 +574,7 @@
               | ord => ord)
           | ord => ord)
       | (Tip_Op t :: rest, Tip_Op t' :: rest') =>
-          (case Lexicon.token_ord (t, t') of
+          (case Lexicon.token_content_ord (t, t') of
             EQUAL => tree_ops_ord (rest, rest')
           | ord => ord)
       | (Node_Op _ :: _, Tip_Op _ :: _) => LESS