fix: Do not use "test"

app.py

@@ -681,7 +681,7 @@ def fetch_results() -> dict[Language, pd.DataFrame]:
                         f"test_{dataset.task.metric}",
                         test_score_dict.get(dataset.task.metric)
                     )
-                    for test_score_dict in record["results"]["raw"]["test"]
+                    for test_score_dict in record["results"]["raw"]
                 ]
                 if dataset.task in data_dict[model_name]:
                     data_dict[model_name][dataset.task].append(scores)