Improvements

- [x] Opentelemetry Grafana Dashboard
- [x] Log processing with otel collector (inject container details etc)
- [ ] Sampling rates (In application and in collector)
- [ ] Basic Alert configuration
- [ ] Crash reporting (sentry)

- [ ] High performance endpoints
- [ ] API for injecting errors
- [ ] Showing how things fail (too many events, too much data, too many dimensions)

Consider:
* UI with end to end tracing
* Dynamic configuration - change filtering levels etc

No file, test, or entry point is named. First clarify which unchecked telemetry item is in scope, then inspect the relevant Rust application and collector configuration; done should be defined as completing one selected item with a verifiable demonstration.

Vom Indexierungsmodell aus dem Issue-Text verfasst.