docs: define M9 production observability strategy

[TechAaroorian]

Summary

• expand M9 from production validation to production observability and validation
• cover both frontend/UI and backend/API failures
• define an open-source monitoring stack:
  • GlitchTip for browser and server error monitoring
  • Uptime Kuma for independent availability checks
  • Playwright for scheduled critical-journey testing
  • GitHub Actions for scheduling and failure artifacts
  • OpenTelemetry as a future server-tracing option
• add release identifiers, correlation IDs, private source maps, and telemetry scrubbing
• define alerting responsibilities and a staged delivery sequence
• add completion evidence for deliberately tested failures and recovery

Why

React error boundaries help the UI recover, but they do not notify the team or capture every browser failure. Server logs also cannot see all errors occurring inside users’ browsers.

M9 now defines a layered strategy for detecting:

• frontend JavaScript and React failures
• Server Action, route, and API errors
• complete application outages
• critical workflows that load but no longer function correctly

Privacy and reliability

The plan requires sensitive values—including credentials, cookies, tokens, and personal form data—to be removed before telemetry is transmitted.

Uptime monitoring will run independently of the SlotSyncro deployment so it can still report an outage when the application infrastructure is unavailable.

Validation

• git diff --check passed
• documentation-only change; application tests and builds were not required

[github-actions]

Coverage Report

Status	Category	Percentage	Covered / Total
🔵	Lines	99.18%	121 / 122
🔵	Statements	98.38%	122 / 124
🔵	Functions	100%	18 / 18
🔵	Branches	84.28%	59 / 70

File Coverage

No changed files found.

Generated in workflow #36 for commit 741be8a by the Vitest Coverage Report Action