fix(langchain): wrap tools with StructuredTool before bind_tools

Raw Python callables passed to bind_tools() lack the schema metadata
required to generate a valid OpenAI tools[].function spec, causing a
400 "Missing required parameter: 'tools[0].function'" error.

Wrapping each callable with StructuredTool.from_function(name=config_key)
also ensures the model response carries the config key as the tool name,
fixing a separate bug where function.__name__ was being tracked instead
of the LaunchDarkly config key.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Author: jsonbailey

packages/ai-providers/server-ai-langchain/src/ldai_langchain/langgraph_agent_graph_runner.py

@@ -53,6 +53,7 @@ async def run(self, input: Any) -> AgentGraphResult:
         start_ns = time.perf_counter_ns()
         try:
             from langchain_core.messages import AnyMessage, HumanMessage
+            from langchain_core.tools import StructuredTool
             from langgraph.graph import END, START, StateGraph
             from typing_extensions import TypedDict
 
@@ -74,11 +75,18 @@ def handle_traversal(node: AgentGraphNode, ctx: dict) -> None:
                 if node_config.model:
                     lc_model = create_langchain_model(node_config)
                     tool_defs = node_config.model.get_parameter('tools') or []
-                    tool_fns = [
-                        tools_ref[t.get('name', '')]
-                        for t in tool_defs
-                        if t.get('name', '') in tools_ref
-                    ]
+                    tool_fns = []
+                    for t in tool_defs:
+                        config_key = t.get('name', '')
+                        if config_key not in tools_ref:
+                            continue
+                        tool_fns.append(
+                            StructuredTool.from_function(
+                                func=tools_ref[config_key],
+                                name=config_key,
+                                description=t.get('description', ''),
+                            )
+                        )
                     model = lc_model.bind_tools(tool_fns) if tool_fns else lc_model
 
                 def invoke(state: WorkflowState) -> WorkflowState: