clean: only attempt to clean platforms which have been installed to

[oliver-sanders]

Addresses the "broken platform" side of #4935.

• Store the remote init map in the workflow database.
• This table is not in itself housekept because the database as a whole is housekept by cylc clean.
• Add a new remote-init status for platforms which have no hosts available (therefor do not need cleaning as we haven't managed to install anything onto them in the first place).

TODO:

• Don't update the status of an install_target if the new status would not require clean and the old would (e.g. can happen on restart if no hosts are available but the remote-init completed for the previous start)
• Requires (extremely simple) DB upgrader (i.e. add blank table on restart).
• Add test.

Requirements 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.

• Appropriate tests are included (unit and/or functional).
• Already covered by existing tests.
• Does not need tests (why?).

• Appropriate change log entry included.
• No change log entry required (why? e.g. invisible to users).

• (master branch) I have opened a documentation PR at cylc/cylc-doc/pull/XXXX.
• (7.8.x branch) I have updated the documentation in this PR branch.
• No documentation update required.

[oliver-sanders]

@hjoliver Does this approach make sense to you?

[oliver-sanders]

Bumping to 8.x as I think this only really impacts us developers dealing with test battery workflows?

For real platforms users will just wait until the platform becomes available again and retry the clean command.

[hjoliver]

Yes, it seems an eminently sensible approach!

[oliver-sanders]

May come back to this one day, but not today...