fix: Use valid CallTypes enum value in embeddings endpoint (#16328)

* Fix embeddings endpoint call_type to use valid CallTypes enum value

Fixed bug where the `/embeddings` endpoint was passing `call_type="embeddings"`
to guardrail hooks, but "embeddings" is not a valid value in the CallTypes enum.

Changed to use `call_type="aembedding"` (async embedding) which is the correct
CallTypes enum value and matches the route_type used in the same function.

Added unit tests to verify:
- "embeddings" is not a valid CallTypes enum value
- "aembedding" is the correct valid value
- The fix prevents ValueError when guardrails are enabled

Fixes #16240

* Inline embeddings call type regression check

* Ensure embedding test preserves proxy metadata

litellm/proxy/proxy_server.py

@@ -5081,7 +5081,7 @@ async def embeddings(  # noqa: PLR0915
 
         ### CALL HOOKS ### - modify incoming data / reject request before calling the model
         data = await proxy_logging_obj.pre_call_hook(
-            user_api_key_dict=user_api_key_dict, data=data, call_type="embeddings"
+            user_api_key_dict=user_api_key_dict, data=data, call_type="aembedding"
         )
 
         tasks = []
@@ -5089,7 +5089,7 @@ async def embeddings(  # noqa: PLR0915
             proxy_logging_obj.during_call_hook(
                 data=data,
                 user_api_key_dict=user_api_key_dict,
-                call_type="embeddings",
+                call_type="aembedding",
             )
         )
 

tests/proxy_unit_tests/test_proxy_server.py

@@ -545,7 +545,27 @@ def test_embedding(mock_aembedding, client_no_auth):
             "input": ["good morning from litellm"],
         }
 
-        response = client_no_auth.post("/v1/embeddings", json=test_data)
+        pre_call_return_value = {
+            **test_data,
+            "metadata": {"source": "unit-test"},
+            "proxy_server_request": {"path": "/v1/embeddings"},
+            "secret_fields": [],
+        }
+
+        with patch.object(
+            litellm.proxy.proxy_server.proxy_logging_obj,
+            "pre_call_hook",
+            new=AsyncMock(return_value=pre_call_return_value),
+        ) as mock_pre_call_hook, patch.object(
+            litellm.proxy.proxy_server.proxy_logging_obj,
+            "during_call_hook",
+            new=AsyncMock(return_value=None),
+        ) as mock_during_hook, patch.object(
+            litellm.proxy.proxy_server.proxy_logging_obj,
+            "post_call_success_hook",
+            new=AsyncMock(return_value=None),
+        ):
+            response = client_no_auth.post("/v1/embeddings", json=test_data)
 
         mock_aembedding.assert_called_once_with(
             model="azure/text-embedding-ada-002",
@@ -559,6 +579,16 @@ def test_embedding(mock_aembedding, client_no_auth):
         result = response.json()
         print(len(result["data"][0]["embedding"]))
         assert len(result["data"][0]["embedding"]) > 10  # this usually has len==1536 so
+
+        pre_call_kwargs = mock_pre_call_hook.await_args_list[0].kwargs
+        assert (
+            pre_call_kwargs.get("call_type") == "aembedding"
+        ), f"expected pre_call_hook to receive call_type='aembedding', got {pre_call_kwargs.get('call_type')}"
+
+        during_call_kwargs = mock_during_hook.await_args_list[0].kwargs
+        assert (
+            during_call_kwargs.get("call_type") == "aembedding"
+        ), f"expected during_call_hook to receive call_type='aembedding', got {during_call_kwargs.get('call_type')}"
     except Exception as e:
         pytest.fail(f"LiteLLM Proxy test failed. Exception - {str(e)}")