scheduler: Add a runner for first visits
This introduces a new runner for scheduling first visits. This matches the current behavior of other runners we have (one runner for listers [runner], another runner for the save-code-now [runner-priority]).
The recently opened cli swh origin schedule-high-priority-first-visits [1] was not
looping. It was doing one round of first visit origins scheduling and then stop. Which,
in a current production-like environment is not enough. This creates pods that crash
loops.
There was no issue in the docker environment as the loop is implemented outside the pre-existing cli. This kept that cli to avoid breaking the docker environment.
This got noticed when trying to review (and validate in a kube environment) the new bulk save-code-now deployment [2]
[1] !390 (merged)
[2] swh/infra/ci-cd/swh-charts!479 (merged)
Refs. #4687