core[minor], integrations...[patch]: Support ToolCall as Tool input and ToolMessage as Tool output (#24038)

Changes:
- ToolCall, InvalidToolCall and ToolCallChunk can all accept a "type"
parameter now
- LLM integration packages add "type" to all the above
- Tool supports ToolCall inputs that have "type" specified
- Tool outputs ToolMessage when a ToolCall is passed as input
- Tools can separately specify ToolMessage.content and
ToolMessage.raw_output
- Tools emit events for validation errors (using on_tool_error and
on_tool_end)

Example:
```python
@tool("structured_api", response_format="content_and_raw_output")
def _mock_structured_tool_with_raw_output(
    arg1: int, arg2: bool, arg3: Optional[dict] = None
) -> Tuple[str, dict]:
    """A Structured Tool"""
    return f"{arg1} {arg2}", {"arg1": arg1, "arg2": arg2, "arg3": arg3}


def test_tool_call_input_tool_message_with_raw_output() -> None:
    tool_call: Dict = {
        "name": "structured_api",
        "args": {"arg1": 1, "arg2": True, "arg3": {"img": "base64string..."}},
        "id": "123",
        "type": "tool_call",
    }
    expected = ToolMessage("1 True", raw_output=tool_call["args"], tool_call_id="123")
    tool = _mock_structured_tool_with_raw_output
    actual = tool.invoke(tool_call)
    assert actual == expected

    tool_call.pop("type")
    with pytest.raises(ValidationError):
        tool.invoke(tool_call)

    actual_content = tool.invoke(tool_call["args"])
    assert actual_content == expected.content
```

---------

Co-authored-by: Erick Friis <erick@langchain.dev>

libs/partners/anthropic/langchain_anthropic/chat_models.py

@@ -43,6 +43,7 @@ from langchain_core.messages import (
     ToolMessage,
 )
 from langchain_core.messages.ai import UsageMetadata
+from langchain_core.messages.tool import tool_call_chunk as create_tool_call_chunk
 from langchain_core.output_parsers import (
     JsonOutputKeyToolsParser,
     PydanticToolsParser,
@@ -1102,12 +1103,12 @@ def _make_message_chunk_from_anthropic_event(
             warnings.warn("Received unexpected tool content block.")
         content_block = event.content_block.model_dump()
         content_block["index"] = event.index
-        tool_call_chunk = {
-            "index": event.index,
-            "id": event.content_block.id,
-            "name": event.content_block.name,
-            "args": "",
-        }
+        tool_call_chunk = create_tool_call_chunk(
+            index=event.index,
+            id=event.content_block.id,
+            name=event.content_block.name,
+            args="",
+        )
         message_chunk = AIMessageChunk(
             content=[content_block],
             tool_call_chunks=[tool_call_chunk],  # type: ignore

libs/partners/anthropic/langchain_anthropic/output_parsers.py

@@ -1,6 +1,7 @@
 from typing import Any, List, Optional, Type, Union, cast
 
 from langchain_core.messages import AIMessage, ToolCall
+from langchain_core.messages.tool import tool_call
 from langchain_core.output_parsers import BaseGenerationOutputParser
 from langchain_core.outputs import ChatGeneration, Generation
 from langchain_core.pydantic_v1 import BaseModel
@@ -79,7 +80,7 @@ def extract_tool_calls(content: Union[str, List[Union[str, dict]]]) -> List[Tool
             if block["type"] != "tool_use":
                 continue
             tool_calls.append(
-                ToolCall(name=block["name"], args=block["input"], id=block["id"])
+                tool_call(name=block["name"], args=block["input"], id=block["id"])
             )
         return tool_calls
     else: