[ARCHIVED] Set up sysctls when tuning nodes

[tnozicka]

What should the feature do:
We should tune sysctls related to scylla on the scylla nodes. Ideally we need to keep a list in sync with what's in regular scylla deployment but enforcing that may not be realistic. Also some values may need to be raised as we could run multiple scylla pods per kubernetes node.

What is use case behind this feature:
Some sysctls can be set directly on our CR but only those that are namespaced (https://kubernetes.io/docs/tasks/administer-cluster/sysctl-cluster/#setting-sysctls-for-a-pod).
fs.inotify.max_user_instances is one of the cases that aren't namespaced and scylla needs them scylladb/scylladb#7700

Additional Information:
https://scylladb-users.slack.com/archives/CCC9J41FW/p1628195109003700

### Tasks
- [ ] https://github.com/scylladb/scylla-operator/issues/1372
- [ ] https://github.com/scylladb/scylla-operator/issues/1373
- [ ] https://github.com/scylladb/scylla-operator/issues/1385

[mykaul]

Feasible for 1.9?

[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