`cylc stop` with cycle point does not work properly when multiple workflows specified

[MetRonnie]

cylc stop a/run1 a/run2 //42

• Expected: both workflows stop after cycle point 42 has completed
• Actual: run1 stops immediately, run2 stops after cycle 42 has completed

cylc stop a/run1//42 a/run2//42

• Expected: same as above
• Actual: InputError: IDs contain too many cycles/tasks/jobs (max 1)

[oliver-sanders]

cylc stop a/run1 a/run2 //42

Expected: both workflows stop after cycle point 42 has completed

I think this is by design but plz check docs/code to check (this is outside of brain cache).

If so, I get why you made that assumption, though examples like this one might explain why it's not implemented that way:

cylc trigger a// //1/x //1/y //1/z  b// //1/i //2/j //3/l

This arrangement imposes fewer constraints on what can be done with the CLI.

---

cylc stop a/run1//42 a/run2//42

Expected: same as above

This should definitely work irrespective, the too many cycles/task/jobscheck is too crude to handle this situation.

Hopefully a simple fix.