Migrate event log to batched logging system

[HyeockJinKim]

Objective

Migrate event log creation from direct DB writes to the batched logging system.

Implementation Details

Event Reporter Changes

1. Update EventLogger (src/ai/backend/manager/event_dispatcher/reporters.py)
   • Replace direct DB write with queue enqueue
   • Batch multiple events together

class EventLogger(AbstractEventReporter):
    def __init__(self, db: ExtendedAsyncSAEngine, queue_manager: LogQueueManager) -> None:
        self._db = db
        self._queue_manager = queue_manager
    
    async def prepare_event_report(
        self, event: AbstractEvent, arg: PrepareEventReportArgs
    ) -> None:
        event_log = EventLogRow.from_event(event)
        await self._queue_manager.enqueue_event_log(event_log)

1. Add Batch Writer for Events
   • Create EventLogBatchWriter
   • Handle bulk insert to event_logs table

Configuration

• Batch size: 200 (higher than other logs, events are frequent)
• Flush interval: 10 seconds (can tolerate delay)
• Redis backup: disabled

Testing

• Test event log enqueueing
• Test high-volume event scenarios
• Integration tests with event dispatcher

Acceptance Criteria

• Event logs written in batches
• No event data loss
• Reduced DB load during high event volume
• All tests passing

JIRA Issue: BA-4236