refactor(graders): Renames the aevaluate method to the internally used _aevaluate.

- Renames the `aevaluate` async method in multiple gradient classes to `_aevaluate` to identify it as an internal method.

- Updates the method call names in `CGCVGrader` to match the renamed `_aevaluate`.

- Maintains asynchronous call logic, enhancing code encapsulation.

- Affected gradients include the `audit`, `cgcv`, `ebtu`, `grounding`, `presentation_quality`, and `traceability` modules.

Author: TaoShuchang

tutorial/example_deep_finance/judge/audit/grader.py

@@ -69,7 +69,7 @@ def create_default_model(
 
         return OpenAIChatModel(**kwargs)
 
-    async def aevaluate(
+    async def _aevaluate(
         self,
         traj: Any,
         **_: Any,

tutorial/example_deep_finance/judge/cgcv/grader.py

@@ -135,7 +135,7 @@ def create_default_model(
 
         return OpenAIChatModel(**kwargs)
 
-    async def aevaluate(
+    async def _aevaluate(
         self,
         traj: Any,
         **_: Any,
@@ -239,7 +239,7 @@ def evaluate(
             loop = asyncio.new_event_loop()
             asyncio.set_event_loop(loop)
 
-        return loop.run_until_complete(self.aevaluate(traj, **kwargs))
+        return loop.run_until_complete(self._aevaluate(traj, **kwargs))
 
     def get_detailed_result(
         self,

tutorial/example_deep_finance/judge/ebtu/grader.py

@@ -48,7 +48,7 @@ def __init__(
             response_format={"type": "json_object"},
         )
 
-    async def aevaluate(self, traj: Any, **kwargs: Any) -> GraderScore:
+    async def _aevaluate(self, traj: Any, **kwargs: Any) -> GraderScore:
         messages = coerce_to_messages_list(traj)
 
         # 输入有效性检查

tutorial/example_deep_finance/judge/grounding/grader.py

@@ -80,7 +80,7 @@ def create_default_model(
 
         return OpenAIChatModel(**kwargs)
 
-    async def aevaluate(
+    async def _aevaluate(
         self,
         traj: Any,
         **_: Any,

tutorial/example_deep_finance/judge/presentation_quality/grader.py

@@ -83,7 +83,7 @@ def create_default_model(
 
         return OpenAIChatModel(**kwargs)
 
-    async def aevaluate(
+    async def _aevaluate(
         self,
         report_content: str,
         user_query: str | None = None,

tutorial/example_deep_finance/judge/traceability/grader.py

@@ -36,7 +36,7 @@ def __init__(
         super().__init__(name=name, **kwargs)
         self.model = model 
 
-    async def aevaluate(self, traj: Any, **kwargs: Any) -> GraderScore:
+    async def _aevaluate(self, traj: Any, **kwargs: Any) -> GraderScore:
         messages = coerce_to_messages_list(traj)
 
         if not messages: