health-probes: clear stale results on start/restart, fast 10s probing until healthy

- Clear HealthProbe on StartStack/RestartStack so stale unhealthy state
  isn't re-applied by RefreshStatus
- Use 10s probe interval for unhealthy/new stacks (nil HealthProbe probes
  immediately on next tick), switch to normal 5m interval once healthy
- Scheduler frequency 1m → 10s to support fast probing

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

controller/internal/stacks/healthprobe.go

@@ -33,10 +33,18 @@ func (m *Manager) RunHealthProbes() error {
 			continue
 		}
 
-		// Check if interval has elapsed since last probe
+		// Check if interval has elapsed since last probe.
+		// Fast 10s probes until healthy, then normal interval (default 5m).
+		// When HealthProbe is nil (just started/restarted), probe immediately.
 		interval := parseInterval(hc.Interval)
-		if stack.HealthProbe != nil && time.Since(stack.HealthProbe.LastCheck) < interval {
-			continue
+		if stack.HealthProbe != nil {
+			effectiveInterval := interval
+			if !stack.HealthProbe.Healthy {
+				effectiveInterval = 10 * time.Second
+			}
+			if time.Since(stack.HealthProbe.LastCheck) < effectiveInterval {
+				continue
+			}
 		}
 
 		// Find the main container to probe (matching stack name)