feat(keeper): implement automated disaster recovery and multi-region failover

[d3vobed]

Summary

Implements an automated disaster recovery and failover system for the keeper service with multi-region RPC endpoint support.

What was added

• New MultiRegionRPCClient failover layer:
  • active endpoint routing
  • automatic endpoint fallback on RPC failure
  • endpoint quarantine with configurable cooldown
  • background health checks and endpoint recovery
• Keeper startup integration in keeper/index.js:
  • uses multi-region failover client as the primary RPC abstraction
  • exposes live failover state to metrics/health
  • lifecycle shutdown handling for failover manager
• Configuration additions in keeper/src/config.js:
  • SOROBAN_RPC_URLS
  • RPC_FAILOVER_ENABLED
  • RPC_FAILOVER_FAILURE_THRESHOLD
  • RPC_FAILOVER_COOLDOWN_MS
  • RPC_FAILOVER_HEALTH_CHECK_INTERVAL_MS
• Observability enhancements in keeper/src/metrics.js:
  • failover counters and gauges
  • failover state in /health and /metrics
  • Prometheus failover metrics
• Documentation:
  • keeper/docs/disaster-recovery-failover.md
  • updates in keeper/README.md and keeper/.env.example
• Tests:
  • keeper/__tests__/disasterRecovery.test.js
  • extended keeper/__tests__/metrics.test.js

Acceptance criteria mapping

• Feature implementation: ✅
• Error tracking/fallback behavior: ✅
• Infrastructure integration: ✅
• Documentation: ✅
• Unit coverage for failover paths: ✅

Validation notes

• Static diagnostics reported no file errors for modified files.
• Runtime tests could not be executed in this container because Node/npm are unavailable and package installation is not permitted in this environment.

Closes #452

[drips-wave]

@d3vobed Great news! 🎉 Based on an automated assessment of this PR, the linked Wave issue(s) no longer count against your application limits.

You can now already apply to more issues while waiting for a review of this PR. Keep up the great work! 🚀

Learn more about application limits