Design LangGraph Workflow for Code Refactoring

Your first task is to design the LangGraph workflow. Define the states, nodes (agents), and edges that will orchestrate the code refactoring process. Include at least three distinct agent roles (e.g., `CodeAnalyzer`, `RefactorProposer`, `TestGenerator`) and define the conditional logic for state transitions. Provide the initial LangGraph definition in Python, including the `StateGraph` and `add_node`/`add_edge` calls. ```python
from typing import TypedDict, Annotated, List
from langchain_core.messages import BaseMessage
from langgraph.graph import StateGraph, END # Define the agent state
class AgentState(TypedDict): code: str refactoring_suggestions: List[str] test_results: str next_action: str # Build the graph
workflow = StateGraph(AgentState) # Add your nodes (agents) and edges here...
# workflow.add_node("analyze", code_analyzer_node)
# workflow.add_edge("analyze", "propose_refactoring")
```