[controller] Astef prototype enlarge our rsc controller

[IvanOgurchenok]

Description

Add ReplicatedVolumeOperation (RVO) API and rolling operations support to RSC controller.

API changes:

• New cluster-scoped ReplicatedVolumeOperation CRD for tracking rolling updates
• New fields in ReplicatedStorageClassStatus: SortSalt, RollingOperationsCursor
• New LastOperation field in ReplicatedVolumeStatus for UI observability

Controller changes:

• RSC controller now creates RVO objects for volumes that need configuration updates or eligible nodes conflict resolution
• Round-robin cursor-based selection ensures fair distribution across volumes
• Budget limits (maxParallel) control concurrent operations
• Automatic cleanup of completed operations when cursor advances

Does not restart any critical cluster components.

Why do we need it, and what problem does it solve?

When ReplicatedStorageClass configuration changes (e.g., topology, replication mode) or eligible nodes list changes, existing volumes need to be updated to match the new configuration.

Previously, there was no mechanism to:

1. Track which volumes need updates
2. Control the rollout pace (avoid updating all volumes at once)
3. Provide visibility into the rollout progress

This PR adds the foundation for controlled rolling updates with:

• RollingUpdate strategy — gradually apply changes to existing volumes
• RollingRepair strategy — automatically fix eligible nodes conflicts
• Configurable parallelism via maxParallel setting

What is the expected result?

After applying:

1. New CRD available: kubectl get crd replicatedvolumeoperations.storage.deckhouse.io

2. RVO objects created when RSC has ConfigurationRolloutStrategy: RollingUpdate or EligibleNodesConflictResolutionStrategy: RollingRepair and volumes need updates: kubectl get rvo

3. RSC status shows rolling state:

   • status.sortSalt — random value for deterministic ordering
   • status.rollingOperationsCursor — tracks progress

4. Conditions reflect rollout status:

   • ConfigurationRolledOut: False with reason ConfigurationRolloutInProgress
   • VolumesSatisfyEligibleNodes: False with reason ConflictResolutionInProgress

Note: RVO objects are created but not processed yet — a separate RVO controller (future PR) will handle the actual volume updates.

Checklist

• The code is covered by unit tests.
• e2e tests passed.
• Documentation updated according to the changes.
• Changes were tested in the Kubernetes cluster manually.