Let the camel jump harness gate on its non-diagnostic rows only

The corpus declares diag- rows as recorded diagnostics of separate
differences (the /ride mount event, RootVehicle); the PR itself reports 6
of them still differing, so exiting 1 on any difference made the new
harness fail on every automatic parity run. They stay in the report.

Co-Authored-By: Claude Opus 5.5 <[email protected]>

tools/test_camel_jump_parity.py

@@ -49,5 +49,7 @@ if __name__ == '__main__':
                   'differences': differences}
         (build / 'strict-report.json').write_text(json.dumps(report, indent=2) + '\n')
         print(f'Strict raw records: {report["passed"]}/{report["records"]}')
-        if differences:
-            raise SystemExit(1)
+        # diag- rows are the corpus's declared diagnostics (separate differences): reported, not gating.
+        failures = [d for d in differences if not d['label'].startswith('diag-')]
+        if failures:
+            raise SystemExit(f'{len(failures)} non-diagnostic rows differ')