docs: Add volume-level snapshot workflow walkthrough

[Copilot]

CloudStack's volume snapshot orchestration spans many layers and classes with non-obvious branching — this adds a single reference document that traces the full path from API call to storage execution.

What's added

docs/volume-snapshot-workflow.md — 11-step sequential walkthrough of the snapshot lifecycle:

• API → Allocation: CreateSnapshotCmd two-phase async create/execute; allocSnapshot() sets up DB record and resource limits
• Validation & dispatch: takeSnapshotInternal() resolves policies, zones, pool IDs; branches on VM attachment — attached volumes serialize through VmWorkTakeVolumeSnapshot job queue, detached volumes execute directly
• Orchestration: orchestrateTakeVolumeSnapshot() builds CreateSnapshotPayload; special-cases live encrypted volumes on StorPool primary storage
• Strategy selection: SnapshotManagerImpl queries StorageStrategyFactory — priority-ordered across StorPoolSnapshotStrategy, StorageSystemSnapshotStrategy, DefaultSnapshotStrategy, Ceph, ScaleIO
• Primary storage execution: SnapshotServiceImpl drives async PrimaryDataStoreDriver.takeSnapshot() via AsyncCallFuture; handles volume state transitions
• Backup path: configurable sync/async secondary backup; StorPool cross-zone copy via native export → remote-copy → reconcile cycle with partial-failure recovery persisted in snapshot_details
• Post-processing: snapshot policy retention, zone/pool replication, delta size resource limit correction, usage events
• Rollback: resource limit rollback on any exception; async backup retry exhaustion triggers cleanupOnSnapshotBackupFailure(); StorPool exports persisted under SP_RECOVERED_SNAPSHOT for cleanup recovery

Also includes a text-form sequence diagram, class/role reference table, and key configuration parameters.

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.