Fix thinko in tok_is_keyword(): it was looking at the wrong union variant

of YYSTYPE, and hence returning the wrong answer for cases where a plpgsql
"unreserved keyword" really does conflict with a variable name.  Obviously
I didn't test this enough :-(.  Per bug #5524 from Peter Gagarinov.

src/test/regress/sql/plpgsql.sql

@@ -2916,6 +2916,20 @@ $$ language plpgsql;
 
 select raise_test();
 
+-- check cases where implicit SQLSTATE variable could be confused with
+-- SQLSTATE as a keyword, cf bug #5524
+create or replace function raise_test() returns void as $$
+begin
+  perform 1/0;
+exception
+  when sqlstate '22012' then
+    raise notice using message = sqlstate;
+    raise sqlstate '22012' using message = 'substitute message';
+end;
+$$ language plpgsql;
+
+select raise_test();
+
 drop function raise_test();
 
 -- test CASE statement