feat(workers): Implement input/output mapping for Data Analyst subgraph

backend/src/ea_chatbot/graph/workers/data_analyst/mapping.py

@@ -0,0 +1,39 @@
+from typing import Dict, Any, List
+from langchain_core.messages import HumanMessage, AIMessage
+from ea_chatbot.graph.state import AgentState
+from ea_chatbot.graph.workers.data_analyst.state import WorkerState
+
+def prepare_worker_input(state: AgentState) -> Dict[str, Any]:
+    """Prepare the initial state for the Data Analyst worker."""
+    checklist = state.get("checklist", [])
+    current_step = state.get("current_step", 0)
+    
+    # Get the current task description
+    task_desc = "Analyze data" # Default
+    if 0 <= current_step < len(checklist):
+        task_desc = checklist[current_step].get("task", task_desc)
+    
+    return {
+        "task": task_desc,
+        "messages": [HumanMessage(content=task_desc)], # Start worker loop with the task
+        "vfs_state": dict(state.get("vfs", {})),
+        "iterations": 0,
+        "plots": [],
+        "code": None,
+        "output": None,
+        "error": None,
+        "result": None
+    }
+
+def merge_worker_output(worker_state: WorkerState) -> Dict[str, Any]:
+    """Map worker results back to the global AgentState."""
+    result = worker_state.get("result", "Analysis complete.")
+    
+    # We bubble up the summary as an AI message
+    updates = {
+        "messages": [AIMessage(content=result)],
+        "vfs": worker_state.get("vfs_state", {}),
+        "plots": worker_state.get("plots", [])
+    }
+    
+    return updates