Revert "Fix CI: import scoring from evaluate_runner instead of openra_rl_util"

This reverts commit 50772fa898f5c685e89d08926d8cb201577c1376.

evaluate.py

@@ -37,7 +37,7 @@ from pathlib import Path
 from typing import Any, Dict, List
 from urllib.request import urlopen
 
-from evaluate_runner import compute_composite_score as compute_composite_score_from_games, compute_game_metrics
+from openra_rl_util.rubrics import compute_composite_score_from_games, compute_game_metrics
 
 # Evaluation results file
 RESULTS_FILE = Path(__file__).parent / "data" / "results.csv"

tests/test_evaluate.py

@@ -154,24 +154,24 @@ class TestAppendResults:
         assert len(RESULTS_COLUMNS) == 13
 
 
-class TestScoringSource:
-    """Verify scoring uses the inlined functions from evaluate_runner."""
-
-    def test_evaluate_imports_from_runner(self):
-        """evaluate.py should import scoring from evaluate_runner."""
-        from evaluate import compute_composite_score_from_games
-        from evaluate_runner import compute_composite_score
-        assert compute_composite_score_from_games is compute_composite_score
-
-    def test_compute_game_metrics_from_runner(self):
-        """evaluate.py should import compute_game_metrics from evaluate_runner."""
-        from evaluate import compute_game_metrics
-        from evaluate_runner import compute_game_metrics as runner_fn
-        assert compute_game_metrics is runner_fn
-
-    def test_score_calculation_basic(self):
-        """Inlined compute_composite_score should produce valid scores."""
-        from evaluate_runner import compute_composite_score
-        games = [{"win": True, "kills_cost": 3000, "deaths_cost": 1000, "assets_value": 8000}]
-        score = compute_composite_score(games)
-        assert 0 < score <= 100
+class TestScoringUsesUtil:
+    """Verify scoring uses the single source of truth from openra-rl-util."""
+
+    def test_rubrics_re_exports_util(self):
+        """rubrics.py should re-export from openra_rl_util."""
+        from rubrics import compute_composite_score_from_games
+        from openra_rl_util.rubrics import (
+            compute_composite_score_from_games as util_fn,
+        )
+        assert compute_composite_score_from_games is util_fn
+
+    def test_evaluate_uses_util_scoring(self):
+        """evaluate.py should not have its own compute_composite_score."""
+        import evaluate
+        assert not hasattr(evaluate, "compute_composite_score"), \
+            "evaluate.py should use compute_composite_score_from_games from Util"
+
+    def test_compute_game_metrics_re_exported(self):
+        from rubrics import compute_game_metrics
+        from openra_rl_util.rubrics import compute_game_metrics as util_fn
+        assert compute_game_metrics is util_fn