[Error Handling] Comprehensive error handling and resilience

[OneByJorah]

Error Handling Checklist

• Add global exception handlers for bot and agent server
• Add circuit breaker pattern for agent communications
• Add retry logic with exponential backoff for external calls
• Add timeout handling for all external HTTP/SNMP calls
• Add graceful degradation when agents/SNMP unavailable
• Add structured error responses with error codes
• Add dead letter queue for failed commands
• Add error rate metrics and alerting
• Add graceful shutdown handling (SIGTERM handling)