fix: guard empty rubric text in substring fallback + debug logging

Address reviewer feedback on #6072:
- Guard `if not rubric and normalized_rubric_text:` prevents empty
  judge Property: lines from matching every rubric via substring
- Guard `if ct and` prevents empty rubric keys from matching
- Add logger.debug when substring fallback rescues a match to track
  judge drift in eval logs
- Add test_empty_property_text_does_not_match test case

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: tottenjordan

src/google/adk/evaluation/rubric_based_evaluator.py

@@ -417,14 +417,19 @@ def convert_auto_rater_response_to_score(
       normalized_rubric_text = _normalize_text(rubric_response.property_text)
       rubric = normalized_rubric_to_rubric_map.get(normalized_rubric_text, None)
 
-      if not rubric:
+      if not rubric and normalized_rubric_text:
         candidates = [
             r
             for ct, r in normalized_rubric_to_rubric_map.items()
-            if ct in normalized_rubric_text or normalized_rubric_text in ct
+            if ct and (ct in normalized_rubric_text or normalized_rubric_text in ct)
         ]
         if len(candidates) == 1:
           rubric = candidates[0]
+          logger.debug(
+              "Rubric substring fallback: '%s' → '%s'",
+              rubric_response.property_text[:60],
+              rubric.rubric_id,
+          )
 
       if rubric:
         rubric_scores.append(

tests/unittests/evaluation/test_rubric_based_evaluator.py

@@ -771,3 +771,11 @@ def test_ambiguous_substring_match_rejected(self):
         judge_property_text="Uses tools",
     )
     assert len(result.rubric_scores) == 0
+
+  def test_empty_property_text_does_not_match(self):
+    """Empty judge Property: line must not match via substring fallback."""
+    result = self._build_evaluator_and_score(
+        rubric_texts=["Uses tools correctly"],
+        judge_property_text="",
+    )
+    assert len(result.rubric_scores) == 0