Integrate peer-based retention logic with distributed storage

[veryCrunchy]

Design and implement Rust logic for peer nodes to temporarily retain (pin/cache) encrypted records for other nodes during downtime. This logic should:

• Detect when a peer node goes offline, identify its records for retention.
• Pin/copy relevant records to local storage (using the fetch/pin interfaces).
• Track retention windows and unpin after expiry or when owner returns.
• Accept and verify signed requests for early deletion from owner nodes.
• Provide status info on which peer is retaining which records.

Implementation Suggestions:

• Build a RetentionManager struct/module.
• Integrate with async liveness monitoring (e.g., using tokio).
• Use event-driven approach: on peer-down, trigger retention; on peer-up, cleanup.
• Ensure all actions are signed and auditable.
• Add integration tests for peer-down/up and data lifecycle.

[obiente]

Connected to Huly®: UP-6