Make armadactl kube accept just jobId

[sync-by-unito]

From jobId we can determine jobset/queue quite simply.

This makes it a lot easier to use this command, rather than having to have a command like:

armadactl kube logs --queue queueA --jobSet some-job-set --jobId <job_id>

It can be as simple as

armadactl kube logs <job_id>

┆Issue is synchronized with this Jira Task by Unito

[sync-by-unito]

➤ Stefan Novakovic commented:

JamesMurkin is this really something that is doable? My understanding is that JobID is strictly identified with queue and job set, in other words it is possible to have same JobID value with different queue/jobset pairs.

The current implementation further supports this notion, as it retrieves information about specific job set first:
https://github.com/G-Research/armada/blob/dbe530525cc9dd6fe2ad29b06d4c56fb6d5960f5/cmd/armadactl/cmd/kube.go#L53-L57