[YUNIKORN-2524] add documentation for recovery queue (root.@recovery@)

[ryankert01]

What is this PR for?

the recovery queue is unqueryable directly but we can observe the recovery queue name via app Restful API (ws/v1/partition/%s/application/%s).

Hence, we should write documents for recovery queue. Otherwise, it would be surprise to users when they see the incomprehensible queue and they get nothing from our docs.

This should be a full documentation of the recovery queue and its purpose. I think we overlooked that until now. With the change to set the force flag in the k8shim we now can see the queue.

We should explain the initialisation process and the roll of the queue:

• name @recovery@ and why that name
• the queue is a dynamic queue and will disappear when not used
• no quota, no ACL
• cannot be submitted to directly
• only used during initialisation for already running allocations that are part of applications that fail placement

What type of PR is it?

• - Bug Fix
• - Improvement
• - Feature
• - Documentation
• - Hot Fix
• - Refactoring

Todos

• - Task

What is the Jira issue?

https://issues.apache.org/jira/browse/YUNIKORN-2524

How should this be tested?

Screenshots (if appropriate)

Questions:

• - The licenses files need update.
• - There is breaking changes for older versions.
• - It needs documentation.

[chenyulin0719]

@ryankert01 Thanks for the PR. Just added a feedback in the Jira, if my opionion does make any sense to you and the others reviewers. We can create another PR or Jira to improve it.