fix: 'MajorityVoter.score' when using multi-labels (#1678)

Fixes the issue mentioned in #1628.
Now the MajorityVoter.score works for multi-label cases with a number of labels > 2.

(cherry picked from commit 1c11c10)

src/rubrix/labeling/text_classification/label_models.py

@@ -375,6 +375,7 @@ def score(
             probabilities = self._compute_multi_label_probs(wl_matrix)
 
             annotation, prediction = self._score_multi_label(probabilities)
+            target_names = self._weak_labels.labels
         else:
             if isinstance(tie_break_policy, str):
                 tie_break_policy = TieBreakPolicy(tie_break_policy)
@@ -384,11 +385,12 @@ def score(
             annotation, prediction = self._score_single_label(
                 probabilities, tie_break_policy
             )
+            target_names = self._weak_labels.labels[: annotation.max() + 1]
 
         return classification_report(
             annotation,
             prediction,
-            target_names=self._weak_labels.labels[: annotation.max() + 1],
+            target_names=target_names,
             output_dict=not output_str,
         )
 

tests/labeling/text_classification/test_label_models.py

@@ -320,9 +320,10 @@ def score(self, probabilities, tie_break_policy=None):
             assert probabilities is None
             if wls == "weak_labels":
                 assert tie_break_policy == TieBreakPolicy.ABSTAIN
-            else:
-                assert tie_break_policy is None
-            return np.array([[1, 1], [0, 0]]), np.array([[1, 1], [1, 0]])
+                return np.array([1, 0]), np.array([1, 1])
+
+            assert tie_break_policy is None
+            return np.array([[1, 1, 1], [0, 0, 0]]), np.array([[1, 1, 1], [1, 0, 0]])
 
         single_or_multi = "multi" if wls == "weak_multi_labels" else "single"
         monkeypatch.setattr(