feat: add operator-aware resizing with examples and automation

[jmealo]

Add operator-aware resizing feature that enables pvc-autoresizer to work harmoniously with Kubernetes operators by patching Custom Resources instead of directly modifying PVCs. Includes comprehensive examples, automation, and documentation for CloudNativePG and RabbitMQ operators.

Core Feature:
- CR patching: Patch operator CR fields instead of direct PVC modification
- Security: RBAC validation + JSONPath restricted to /spec/* only
- Metrics: CR patch success/failure counters with operator labels
- Events: Kubernetes events for CR operations (ResizedCR, ResizeCRFailed)

New files:

• internal/runners/cr_patcher.go: Core CR patching logic with validation
• internal/runners/cr_patcher_test.go: Unit tests including JSONPath validation
• docs/operator-aware-resizing.md: Complete documentation with security model, configuration, monitoring, troubleshooting, and supported operators
• examples/kyverno-cnpg-autoresizer.yaml: Production-ready Kyverno policy for automatic CNPG PVC annotation based on role (data, WAL, tablespaces)
• examples/annotate-pvcs.sh: Tested kubectl commands for CloudNativePG, RabbitMQ, and Strimzi Kafka operators
• examples/README.md: Complete testing guide with step-by-step procedures and verified timelines

Testing:
All examples tested and verified on live clusters on Azure AKS:

• CloudNativePG: Data, WAL, and tablespace volumes working end-to-end
• RabbitMQ: Persistence volumes with no reconciliation conflicts
• Kyverno policy: Automatic annotation with dynamic label extraction
• Unit tests: JSONPath validation and security enforcement

Key features:

• Solves reconciliation conflicts with operator-managed PVCs
• Automatic PVC role detection via operator labels (cnpg.io/pvcRole)
• Dynamic tablespace name extraction for JSONPath filters
• Both manual (kubectl) and automated (Kyverno) workflows
• Production-ready configurations with namespace selectors
• Defense-in-depth security (RBAC + code-enforced path validation)

Implements: #346

[jmealo]

@llamerada-jp Happy New Year! Hello, wanted to follow up on this. I've been using this in my clusters and it's been working well for me with CNPG on Azure AKS.