Requeue failed jobs from AsyncResult__c

[Mateusz7410]

Motivation

When a job exhausts its retries and lands in Failed, the only recovery today is
to re-run the originating code by hand. An operational requeue would let a
dev or admin re-enqueue a failed job straight from its AsyncResult__c record,
which matters most after a transient outage leaves a batch of jobs failed.

This is distinct from retry(): retry() is automatic, runs inside the failure
path, and is bounded by maxRetries and backoff. Requeue is manual and runs
after a job is already in terminal Failed, with a fresh attempt counter.

Proposal

• An entry point to re-enqueue a failed job by AsyncResult__c id (or custom job
  id): reconstruct / clone the job and push it back through the normal enqueue
  path with a fresh attempt counter.
• Optional bulk requeue for a set of failed results.
• If the monitoring console ships, expose this as a row action / button there.

Notes

• Requires the job payload to be reconstructable. Document the constraint: either
  the job state is serialized on failure, or requeue is limited to jobs that
  carry their own inputs.
• Guard with a custom permission so not every user can requeue.

Acceptance criteria

• Requeue path plus permission gate.
• Apex tests: single + bulk requeue, non-failed result rejected, missing payload
  handled cleanly.
• Docs.