feat: message aggregator (#1985)

* feat: aggregator

* fix: resolve deadlock, mutation, and safety issues in message aggregator

- Fix deadlock: don't await cancelled timer tasks inside the lock;
  _flush_buffer acquires the same lock, causing a deadlock cycle
- Fix message_event mutation: keep original message_event unmodified
  to preserve message_id/metadata for reply/quote; only pass merged
  message_chain separately
- Fix Plain positional arg: Plain('\n') → Plain(text='\n')
- Fix float() ValueError: wrap delay cast in try/except
- Add MAX_BUFFER_MESSAGES (10) cap to prevent unbounded buffer growth
- Default enabled to false to avoid surprising latency on upgrade
- Fix flush_all: cancel all timers under one lock acquisition, then
  flush outside the lock to avoid deadlock

---------

Co-authored-by: RockChinQ <rockchinq@gmail.com>

src/langbot/pkg/platform/botmgr.py

@@ -82,7 +82,7 @@ class RuntimeBot:
                     if custom_launcher_id:
                         launcher_id = custom_launcher_id
 
-                await self.ap.query_pool.add_query(
+                await self.ap.msg_aggregator.add_message(
                     bot_uuid=self.bot_entity.uuid,
                     launcher_type=provider_session.LauncherTypes.PERSON,
                     launcher_id=launcher_id,
@@ -125,7 +125,7 @@ class RuntimeBot:
                     if custom_launcher_id:
                         launcher_id = custom_launcher_id
 
-                await self.ap.query_pool.add_query(
+                await self.ap.msg_aggregator.add_message(
                     bot_uuid=self.bot_entity.uuid,
                     launcher_type=provider_session.LauncherTypes.GROUP,
                     launcher_id=launcher_id,