feat: Add cgroup v2 support for accurate container resource monitoring

Problem

Modern Linux distributions and containers (Docker/Kubernetes) use cgroup v2. Current prmon monitoring has reduced accuracy in containerized environments and cannot detect container resource limits.

Proposed Solution

Add new cgroupmon monitor component that:

• Auto-detects and reads from cgroup v1/v2 (/sys/fs/cgroup/)
• Provides accurate CPU, memory, I/O stats for containers
• Reports container limits and quotas
• Maintains backward compatibility with non-container environments

Expected Deliverables

• cgroupmon.cpp/h implementing Imonitor interface
• Support for both cgroup v1 and v2
• Auto-detection with fallback to /proc
• Unit tests with precooked sources
• Integration tests in Docker/Kubernetes
• Documentation updates (README, ADDING_MONITORS.md)

Benefits

• Accurate monitoring in modern container platforms
• Lightweight implementation (no new dependencies)
• Suitable for HEP workflows in Kubernetes

Start by reading the existing monitor components and the Imonitor interface before designing cgroupmon.cpp/h. Check how sources under /sys/fs/cgroup/ and /proc are currently handled, then plan unit tests with precooked sources and Docker/Kubernetes integration tests. Done means cgroup v1/v2 detection, limits and CPU, memory, and I/O statistics, fallback behavior, and updates to README and ADDING_MONITORS.md.