fix: GroundednessEvaluator.has_context should be private (#43336)

Author: kdestin

sdk/evaluation/azure-ai-evaluation/azure/ai/evaluation/_evaluators/_groundedness/_groundedness.py

@@ -214,7 +214,7 @@ def __call__(  # pylint: disable=docstring-missing-param
 
         return super().__call__(*args, **kwargs)
 
-    def has_context(self, eval_input: dict) -> bool:
+    def _has_context(self, eval_input: dict) -> bool:
         """
         Return True if eval_input contains a non-empty 'context' field.
         Treats None, empty strings, empty lists, and lists of empty strings as no context.
@@ -235,7 +235,7 @@ async def _do_eval(self, eval_input: Dict) -> Dict[str, Union[float, str]]:
         if "query" not in eval_input:
             return await super()._do_eval(eval_input)
 
-        contains_context = self.has_context(eval_input)
+        contains_context = self._has_context(eval_input)
 
         simplified_query = simplify_messages(eval_input["query"], drop_tool_calls=contains_context)
         simplified_response = simplify_messages(eval_input["response"], drop_tool_calls=False)