failed core jobs should not have follow-ups

[tgross]

Fixes #8658

If an evaluation fails more than the delivery limit, the leader will create a new
eval with the TriggeredBy field set to failed-follow-up.

Evaluations for core jobs have the leader's ACL, which is not valid on another
leader after an election. The failed-follow-up evals do not have ACLs, so
core job evals that fail more than the delivery limit or core job evals that
span leader elections will never succeed and will be re-enqueued forever. So
we should not retry with a failed-follow-up.

---

Failing test before the patch in 025a613:

nomad-613 2020-08-17T20:27:52.184Z [WARN]  nomad/leader.go:641: nomad: eval reached delivery limit, marking as failed: eval="<Eval "1bf00c66-09a5-a3fb-488a-f30ecede9798" JobID: "csi-plugin-gc" Namespace: "-">"
    core_sched_test.go:2452: 
        	Error Trace:	core_sched_test.go:2452
        	Error:      	Expected nil, but got: <Eval "574cf08f-c0ee-b12b-2ef3-d257380bae8e" JobID: "csi-plugin-gc" Namespace: "-">
        	Test:       	TestCoreScheduler_FailLoop
        	Messages:   	failed core jobs should not result in follow-up. TriggeredBy: failed-follow-up

[github-actions]

I'm going to lock this pull request because it has been closed for 120 days ⏳. This helps our maintainers find and focus on the active contributions.
If you have found a problem that seems related to this change, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.