disable chunk info scrape when shard cluster have lots of collections

[naughtyGitCat]

Is your feature request related to a problem? Please describe.
When cluster have 100+ collections(sharded or not sharded), getting collection chunk info took a lot of time, in my situation ( 3 shard instances, 10+ databases, 100+collections) average 10s

Describe the solution you'd like
Add the chunk scrape option to --collector.collstats-limit, when --collector.collstats-limit is set and hit, the chunk info is not scraped

Describe alternatives you've considered
or another flag --collector.collchunkinfo-limit to set this feature

[github-actions]

This issue has been marked as stale because it has been open for 120 days without activity. Please remove the stale label or add a comment; otherwise, it will be closed in 7 days.