feat: Add resource limit monitoring for CPU, memory, and I/O

[gma1k]

Add Resource Limit Monitoring

Summary

Implements resource limit monitoring for Kubernetes pods, tracking CPU, memory, and I/O utilization against cgroup limits with configurable alert thresholds and full integration into Podtrace's diagnostic reporting system.

Features

• Resource Monitoring: Tracks CPU, memory, and I/O usage vs. limits from cgroup filesystems
• Alert System: Configurable thresholds (80% WARNING, 90% CRITICAL, 95% EMERGENCY)
• eBPF Integration: Uses BPF maps (cgroup_limits, cgroup_alerts) for kernel-space data storage
• Diagnostic Reports: Resource limit statistics and alerts included in auto-detection summary
• Prometheus Metrics: Exports podtrace_resource_limit_bytes, podtrace_resource_usage_bytes, podtrace_resource_utilization_percent, and podtrace_resource_alert_level
• Event System: Emits EVENT_RESOURCE_LIMIT events integrated with existing event processing pipeline
• Issue Detection: Automatically flags high resource utilization in diagnostic reports

Testing

• Unit tests for all parsing functions
• Integration tests for event formatting, issue detection, and report generation