fixes #23825; Busy wait on waitid, sleeping at regular intervals

[maleyva1]

Addresses #23825 by using the approaching described in #23743 (comment).

This takes the approach from Python's subprocess library which calls waitid in loop, while sleeping at regular intervals.

CC @alex65536

[Araq]

Irrelevant CI failures.

[github-actions]

Thanks for your hard work on this PR!
The lines below are statistics of the Nim compiler built from 58b36bd

Hint: mm: orc; opt: speed; options: -d:release
173130 lines; 7.963s; 664.164MiB peakmem

[alex65536]

Good job 👍

But I think that there are a few minor issues:

• I'm unsure whether using 1ms as an upper limit is a good idea, I would put the limit closer to 50ms, just as Python does. Though, I am really not sure what is the ideal solution here.
• To calculate the remaining time, getTime() is used. But getTime() is not monotonic and is able to go backwards if either the user or the NTP daemon changes the system time. I think, getMonoTime() from monotimes would be better to compute timeouts.

[Araq]

You're correct.

[maleyva1]

@alex65536 @Araq I made a PR (#23834) addressing these issues.