fix(deepseek): convert tool output arrays to strings (#31913)

## Description
When ChatDeepSeek invokes a tool that returns a list, it results in an
openai.UnprocessableEntityError due to a failure in deserializing the
JSON body.

The root of the problem is that ChatDeepSeek uses BaseChatOpenAI
internally, but the APIs are not identical: OpenAI v1/chat/completions
accepts arrays as tool results, but Deepseek API does not.

As a solution added `_get_request_payload` method to ChatDeepSeek, which
inherits the behavior from BaseChatOpenAI but adds a step to stringify
tool message content in case the content is an array. I also add a unit
test for this.

From the linked issue you can find the full reproducible example the
reporter of the issue provided. After the changes it works as expected.

Source: [Deepseek
docs](https://api-docs.deepseek.com/api/create-chat-completion/)


![image](https://github.com/user-attachments/assets/a59ed3e7-6444-46d1-9dcf-97e40e4e8952)

Source: [OpenAI
docs](https://platform.openai.com/docs/api-reference/chat/create)


![image](https://github.com/user-attachments/assets/728f4fc6-e1a3-4897-b39f-6f1ade07d3dc)


## Issue
Fixes #31394

## Dependencies:
No new dependencies.

## Twitter handle:
Don't have one.

---------

Co-authored-by: Mason Daugherty <[email protected]>
Co-authored-by: Mason Daugherty <[email protected]>

Author: nikk0o046

libs/partners/deepseek/langchain_deepseek/chat_models.py

@@ -2,6 +2,7 @@
 
 from __future__ import annotations
 
+import json
 from collections.abc import Iterator
 from json import JSONDecodeError
 from typing import Any, Literal, Optional, TypeVar, Union
@@ -218,6 +219,19 @@ def validate_environment(self) -> Self:
             self.async_client = self.root_async_client.chat.completions
         return self
 
+    def _get_request_payload(
+        self,
+        input_: LanguageModelInput,
+        *,
+        stop: Optional[list[str]] = None,
+        **kwargs: Any,
+    ) -> dict:
+        payload = super()._get_request_payload(input_, stop=stop, **kwargs)
+        for message in payload["messages"]:
+            if message["role"] == "tool" and isinstance(message["content"], list):
+                message["content"] = json.dumps(message["content"])
+        return payload
+
     def _create_chat_result(
         self,
         response: Union[dict, openai.BaseModel],

libs/partners/deepseek/tests/unit_tests/test_chat_models.py

@@ -5,7 +5,7 @@
 from typing import Any, Literal, Union
 from unittest.mock import MagicMock
 
-from langchain_core.messages import AIMessageChunk
+from langchain_core.messages import AIMessageChunk, ToolMessage
 from langchain_tests.unit_tests import ChatModelUnitTests
 from openai import BaseModel
 from openai.types.chat import ChatCompletionMessage
@@ -217,3 +217,19 @@ def test_convert_chunk_with_empty_delta(self) -> None:
             msg = "Expected chunk_result not to be None"
             raise AssertionError(msg)
         assert chunk_result.message.additional_kwargs.get("reasoning_content") is None
+
+    def test_get_request_payload(self) -> None:
+        """Test that tool message content is converted from list to string."""
+        chat_model = ChatDeepSeek(model="deepseek-chat", api_key=SecretStr("api_key"))
+
+        tool_message = ToolMessage(content=[], tool_call_id="test_id")
+        payload = chat_model._get_request_payload([tool_message])
+        assert payload["messages"][0]["content"] == "[]"
+
+        tool_message = ToolMessage(content=["item1", "item2"], tool_call_id="test_id")
+        payload = chat_model._get_request_payload([tool_message])
+        assert payload["messages"][0]["content"] == '["item1", "item2"]'
+
+        tool_message = ToolMessage(content="test string", tool_call_id="test_id")
+        payload = chat_model._get_request_payload([tool_message])
+        assert payload["messages"][0]["content"] == "test string"