Update execute.py

result dict ['results'] -> ['result'], to stay consistent with code_eval

execute.py

@@ -50,7 +50,7 @@ def check_correctness(program, test_inputs, test_outputs, timeout, task_id, comp
     return dict(
         task_id=task_id,
         passed=all(result.get("status") == "passed" for result in results),
-        results=list(results),
+        result=list(results),
         completion_id=completion_id,
         program=program,
     )