Amend CLI globbing help text.

[hjoliver]

Update CLI globbing help, post #3296

(We need to be very clear that CLI globbing matches against the current task pool, not abstract tasks in the workflow)

(This may need updating again if we implement pseudo-spawn-on-demand with automatic insertion of tasks not in the pool).

---

This is a small change with no associated Issue.

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).
• Does not need tests (CLI help text only).
• Appropriate change log entry included.
• No documentation update required (is documentation).

[sadielbartholomew]

Looks good, apart from one small point, & also see my comment RE the position of the warning from the Cylc-7 twin PR, which also applies here & also in this case to the 'note' for the cycle-point-excluded globbing case (e.g. cylc isnert --help).

[hjoliver]

(I'll reposition the warning message on this branch too).

[sadielbartholomew]

Good clarifications, I think they will help users to understand the real nature of the commands. There are still a few tweaks need here, ideally.

[hjoliver]

Oops - @sadielbartholomew I hadn't requested re-review of the this branch yet (just the 7.8.x) - sorry I should have explicitly said I was still working on it. I've already addressed some of your comments (but not pushed yet). However, I had missed the "TASk" typo - good spotting! Sorry again 😬

[sadielbartholomew]

Ah sorry @hjoliver, I guess when you said one was done I assumed the other was too. My apologies!

[hjoliver]

Now ready for review again.

[sadielbartholomew]

Tested as displaying the new or amended content with a sensible order & formatting for the --help output of various affected commands.

[sadielbartholomew]

Actually hold that for a moment, some of the Travis CI test failures may be legit, so just investigating those.

[sadielbartholomew]

I have doubled checked with Matt who has been dealing with flaky tests, & all Travis CI test failures (see chunk 2/4) are flaky tests which are known to him (one did fail locally for me, but we determinedit was not related to the change here). So they are unrelated, & this is good to go. Thanks.