Fix metrics monitoring

[Devadakene]

Closes #877

Description:
This PR addresses incomplete operational monitoring by introducing granular metrics for charge results and RPC failovers, ensuring improved visibility and production readiness.

Changes:

• scripts/metrics-server.ts: Added the keeper_charge_results_total counter to track detailed individual outcomes (Charged, Skipped, Paused, GracePeriodElapsed, etc.) via a new result label. Added the keeper_rpc_failovers_total counter to track RPC client failover events.
• scripts/keeper.ts: Integrated the metrics-server directly into the keeper execution cycle. Both processPageLive and processPageDryRun now measure duration and report the granular charge outcomes. The startMetricsServer is safely initialized on startup.
• scripts/rpc-client.ts: Instrumented the executeWithFailover loop to increment the failover metric whenever an RPC endpoint fails and the client falls back to an alternative.
• scripts/grafana-dashboard.json: Appended two new panels to track the metrics:
  • Charge Results Breakdown: A timeseries chart tracking granular outcomes per cycle.
  • RPC Failover Rate: A stat panel monitoring RPC endpoint instability and failovers.

Acceptance Criteria Met:

• Metrics endpoint now lists the new series (keeper_charge_results_total, keeper_rpc_failovers_total).
• Dashboard JSON updated to reference the new series.
• Uses low cardinality labels (result mapping exclusively to strict enum variants and string types).
• Validation allows metrics to be gathered and viewed locally during dry keeper cycles.

[drips-wave]

@Devadakene 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