F3: Synthetic Monitoring / Probing Engine

## Summary
Deterministic, scheduled, non-AI probing: checks network connectivity, disk usage, and container/pod health on a schedule. Foundation for both fault detection and future status page.

## Acceptance Criteria
- [ ] Scheduled probes: network connectivity, disk usage, container/pod health
- [ ] Higher-tier plans get more frequent probing (upsell dimension)
- [ ] Probing runs in staging/internal-only patrol mode by Sprint 3
- [ ] Production-ready by Sprint 4
- [ ] AI invoked only on anomaly detection → produces summarized conclusion
- [ ] Natural foundation for public-facing status page (backlog)

## Sprint Assignment
- Sprint 1: Design starts
- Sprint 2–3: Build continues
- Sprint 3: Running in staging/internal patrol mode
- Sprint 4: Production launch with diagnosis layer

## Success Metrics
- Probing coverage (% of active projects covered)
- False positive rate
- Anomaly detection latency

The issue names no files, tests, or entry points. Start by locating the existing scheduling, health-check, container/pod, and plan-tier components, then clarify the design and rollout boundaries. Done requires scheduled probes, staging patrol operation, production readiness, and anomaly-triggered conclusions matching the acceptance criteria.

Written by the indexing model from the issue text.