[Docs] Add more info on Snapshot Restore process

[stefnestor]

@jrodewig / team - I'd like to recommend adding into docs the two most common methods of using your docs to restore snapshots (delete and restore, or restore side-by-side and then delete).

Outlined here: https://medium.com/@stefnestor/tldr-elasticsearch-snapshot-restore-7b72dd75d74a

(It does include some non-documented but working Elasticsearch URLs, hope that's ok.)

Thanks for the great work you do!

[elasticmachine]

Pinging @elastic/es-docs (Team:Docs)

[jrodewig]

(It does include some non-documented but working Elasticsearch URLs, hope that's ok.)

That's a-okay. That's likely a content gap on our end. Thanks for raising this @stefnestor!

[elasticmachine]

Pinging @elastic/es-distributed (Team:Distributed)

[stefnestor]

Also for full cluster restores, I've been recommended to turn off Kibana, ML, ILM, & SLM.

• ILM
• Kibana: stop service on local machine, in ESS "stop routing requests" (not fully necessary maybe unless you have users doing heavy search requests)
• SLM: start & stop
• ML

[jrodewig]

It looks like we already document full cluster restores in Back up a cluster. However, we don't mention stopping ILM, Kibana, SLM, or ML. We could probably also streamline those docs a bit and cross-link from the snapshot and restore section.

[jrodewig]

#72497 covers how to restore an index/data stream from a snapshot.

I'll raise another PR for the full cluster restore docs once that's merged.