Remove Rubix caching

[wendigo]

Rubix was abandoned 3 years ago, it's broken in majority of production cases with frequent cluster crashes, it has number of high impact CVEs and we are planning to replace it with Alluxio caching. Keeping Rubix in its current form brings more harm than good.

Description

Additional context and related issues

Release notes

( ) This is not user-visible or is docs only, and no release notes are required.
( ) Release notes are required. Please propose a release note for me.
(x) Release notes are required, with the following suggested text:

## Hive connector
* Remove object storage caching support. ({issue}`20102`)

[mosabua]

Ideally we merge the replacement Alluxio caching PR first .. but I am also fine merging this now. At minimum however we need to also remove the documentation.

[wendigo]

@mosabua Rubix and Alluxio are unpatched and unmaintained dependencies that have high score CVEs. Rubix doesn't work so keeping it doesn't bring any value other than false impression that hive caching is actually supported. Yeah, I'd love to see new Alluxio land as it has a higher chance of being maintained, supported and actually working.

stale

[mosabua]

Lets ship it!

[jkylling]

Thank you!

[mosabua]

Imho we should merge - provided the CI failures are false alarms.

[martint]

We’re going to merge it after the Alluxio changes get in.

[mosabua]

Sounds good @martint - once we have coverage with Alluxio caching for Delta Lake from the current PR or once we have it also for Hive?

[wendigo]

Rebased.

[wendigo]

Rebased