Split operator auxiliary processes out of ScyllaDB container and remove the need to copy the operator binary there

[tnozicka]

We should avoid running any operator logic in ScyllaDB container, as it's bad design and hurts ScyllaDB latency.

This will also effectively remove the hack where we copy the scylla-operator binary into the ScyllaDB container which takes a lot of time (init container has low cpu allocation) and significantly slows down the node start up time.

Tasks

Preview Give feedback

1. Split scylla probes into a separate container #620
   kind/epic priority/important-longterm +2
   
2. Split startup logic out of ScyllaDB container #1941
   kind/feature priority/important-soon triage/accepted +3
   
3. Split operator sidecar out of ScyllaDB container #1942
   kind/feature priority/important-soon triage/accepted +3
   
4. Move ScyllaDB rack properties generation out of the sidecar #1720
   kind/cleanup priority/important-soon triage/accepted +3
   
5. Move scylla config creation into the operator #604
   kind/epic priority/important-longterm triage/accepted +3
   
6. Move node-exporter to its own container #2290
   kind/feature priority/important-longterm +2

[scylla-operator-bot]

The Scylla Operator project currently lacks enough contributors to adequately respond to all issues.

This bot triages un-triaged issues according to the following rules:

• After 30d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out

/lifecycle stale

[tnozicka]

/remove-lifecycle stale
/triage accepted