fix(backup-dashboard): show last SUCCESSFUL backup + last attempt separately

The dashboard widget used `lastBackup.created_at` for the "Last
successful backup: X ago" text — but lastBackup is the most recent
row of any status. So a crashed restore (status=running, never
updated) or a recent failure showed up labeled as the last
successful backup. Same "silent failure not surfaced" class the
restore wizard had.

Backend now returns:
  lastSuccessfulBackup — most recent backup_runs with status='completed'
  zombieRuns — running rows older than 30 min (likely crashed mid-flight)
  lastBackup — unchanged (most recent any status)

Frontend renders:
  - "Last successful backup: X ago"  — always from lastSuccessfulBackup
  - "Last attempt: Y ago · failed/running"  — when lastBackup differs
    from lastSuccessful. failed shows the first line of error_message
    in red; running stays neutral.
  - Zombie callout — "N backup(s) running >30min — may have crashed"
    in amber, so admin sees stuck rows at a glance.
  - Health score downgrades from "excellent" to "warning" if the
    latest attempt failed, even when older successes keep the age
    fresh — surfaces regressions without erasing the green history.

backend/src/services/backupService.js

@@ -1237,11 +1237,34 @@ async function getBackupStatus(limit = 10) {
 
     const lastRunWithManifest = lastRun ? { ...lastRun, manifestValid } : null;
 
+    // Separate "most recent attempt" from "most recent SUCCESS" so the
+    // dashboard widget can distinguish:
+    //   - last attempt failed → red, "Last attempt failed at X"
+    //   - last attempt running → blue spinner, "In progress since X"
+    //   - never succeeded → critical, "No successful backup yet"
+    //   - last attempt succeeded → green tick, "Last backup X ago"
+    // Previously the widget showed the most-recent row with a generic
+    // green tick regardless of status, so a crashed run from 5 minutes
+    // ago looked identical to a successful one. Same "silent failure
+    // not surfaced" class Stage A was designed to fight.
+    const lastSuccessful = runs.find(r => r.status === 'completed') || null;
+    // Detect zombie running rows (started >30min ago, never updated)
+    // — these are processes that died without writing a completed_at.
+    // Surface them so the admin can tell at a glance vs a live run.
+    const ZOMBIE_THRESHOLD_MS = 30 * 60 * 1000;
+    const zombieRuns = runs.filter(r =>
+      r.status === 'running'
+      && r.started_at
+      && (Date.now() - new Date(r.started_at).getTime()) > ZOMBIE_THRESHOLD_MS
+    );
+
     return {
       isRunning,
       isHealthy: Boolean(lastRun && lastRun.status === 'completed'),
       lastRun: lastRunWithManifest,
       lastBackup: lastRunWithManifest, // Alias for frontend compatibility
+      lastSuccessfulBackup: lastSuccessful, // NEW — see comment above
+      zombieRuns,                           // NEW — running >30min, likely crashed
       recentRuns: runs,
       recentBackups: runs, // Alias for frontend compatibility
       totalBackups: runs.filter(r => r.status === 'completed').length,