Don't demote job failed messages to debug.

[hjoliver]

Currently, we demote task messages to DEBUG level if they are more or less redundant with logged state changes.

However, this isn't ideal for job failure messages because,

• if there are any retries lined up, the state change straight to waiting, with no indication of failure, looks wrong
• job failures should really always be logged as critical (even if the graph is designed to handle it)

Example:

[scheduling]
    [[graph]]
        R1 = "a"
[runtime]
    [[a]]
        script = false
        execution retry delays = PT5S

Current:

This branch:

Check List

• I have read CONTRIBUTING.md and added my name as a Code Contributor.
• Contains logically grouped changes (else tidy your branch by rebase).
• Does not contain off-topic changes (use other PRs for other changes).
• Applied any dependency changes to both setup.cfg (and conda-environment.yml if present).
• Tests are included (or explain why tests are not needed).
• Changelog entry included if this is a change that can affect users
• Cylc-Doc pull request opened if required at cylc/cylc-doc/pull/XXXX.
• If this is a bug fix, PR should be raised against the relevant ?.?.x branch.

[hjoliver]

I don't think this warrants a change log entry or tests.

[oliver-sanders]

Please see #6169

[hjoliver]

Closing as a duplicate of #6169 in light of discussion to change the log level of both submission and execution failures (with retries) in the same PR.