feat: thread-aware memory with /flush and /recall commands

steward/main.py

@@ -8,6 +8,7 @@ from apscheduler.schedulers.asyncio import AsyncIOScheduler
 from steward.bot.telegram import build_application, send_proposal
 from steward.config import get_settings
 from steward.llm.client import LLMClient
+from steward.memory.thread_store import ThreadMemoryStore
 from steward.proposals.generator import ProposalGenerator
 
 logging.basicConfig(
@@ -44,7 +45,8 @@ def main() -> None:
         sys.exit(1)
 
     llm = LLMClient(settings)
-    app = build_application(settings, llm)
+    thread_store = ThreadMemoryStore(settings.thread_memory_path)
+    app = build_application(settings, llm, thread_store)
     generator = ProposalGenerator(settings, llm)
 
     scheduler = AsyncIOScheduler()