Don't recompute runahead unnecessarily in main loop

[hjoliver]

Follow-up #7237 (see last paragraph of #7237 (comment) )

I did not think that calling compute_runahead once per main loop would be a significant efficiency hit in inactive workflows because the method returns early if the base point hasn't changed. So in inactive workflows it really only does this:

base_point = min(self.active_tasks)

HOWEVER, it's certainly better not to call it at all if nothing is happening - so fair call @oliver-sanders and @MetRonnie

Note the change in task_pool.py (commit 2) is not necessary, it's just to separate out the soon-to-be-gone Cylc 7 compat bit, allowing the Cylc 8 case to be cleaner (and avoids a Python loop).

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]

(Changed PR branch base several times after accidentally screwing up 8.6.x in my own fork 😬 )

[hjoliver]

Lint fail unrelated - see #7403