Integration: OWASP Agent Memory Guard for memory security observability

## Integration Request: Memory Security Monitoring

Hi AgentOps team,

AgentOps provides excellent observability for AI agents — session tracking, cost monitoring, error detection. One gap in the current observability stack is **memory security**: detecting when an agent's memory has been poisoned, tampered with, or is leaking secrets.

**OWASP Agent Memory Guard** is the reference implementation for [OWASP ASI06 (Memory Poisoning)](https://owasp.org/www-project-top-10-for-large-language-model-applications/). It could complement AgentOps by adding a security dimension to agent observability:

- Emit security events (injection attempts, tampering detected, secrets found) as AgentOps spans
- Track memory integrity scores alongside cost/latency metrics
- Alert on anomalous memory write patterns across sessions

```python
pip install agent-memory-guard
```

**GitHub:** https://github.com/OWASP/www-project-agent-memory-guard
**PyPI:** https://pypi.org/project/agent-memory-guard/

Would love to explore a native integration or at minimum a "Security Monitoring" guide in the AgentOps docs. Happy to contribute.