Context

Follow-up to WT-18640 for the Single Cluster and Fast Truncate Fleet Health dashboards.

Several truncate metrics currently come from connection-level WiredTiger statistics. These include activity from tables other than the oplog, so the dashboards can attribute unrelated truncate activity to the oplog.

This work depends on SPM-4597 (WiredTiger observability enablement) making the required table-level metrics available through OTel, and WT-18854 correcting layered-table statistics attribution and reporting.

Scope

1. Identify which health-check inputs and supporting measurements should track the oplog specifically, including truncate-list traversal, conflicts, ingest activity, and preceding truncate-list evidence where table-level statistics are available.
2. Replace those connection-level inputs with OTel metrics filtered to the logical oplog table. Resolve the table identity consistently for each cluster/node and preserve it across role changes and restarts.
3. Update the single-cluster and fleet calculations, supporting panels, and required-metric coverage checks consistently.