Quiet loop logs; improve loop intervals

[jlowin]

Thanks for contributing to Prefect Server!

Please describe your work and make sure your PR:

• adds new tests (if appropriate)
• adds a changelog entry to the changes/ directory (if appropriate)

Note that your PR will not be reviewed unless these two boxes are checked.

What does this PR change?

This PR:

• Quiets debug logs from loop services to heartbeat no more than once every 5 minutes. For rapid services, the noise from heartbeating every loop makes working with logs difficult
• Improves the loop interval calculation so it starts from the beginning of each period not the end. If a service takes one hour to run and has a loop interval of 24 hours, this means it will now start every 24 hours instead of every 25 hours.
• Adds a warning if the loop interval is shorter than the execution time (which would lead to too-rapid execution without pauses)

[codecov-commenter]

Codecov Report

Merging #43 into master will increase coverage by 0.23%.
The diff coverage is 88.23%.

[jlowin]

@cicdw made the following changes:

• added a stop() method to facilitate testing and interactive use (otherwise the services run infinitely)
• added tests, including for the situation you described
• removed the random starting delay - overengineered complexity with no clear benefit ✅
• turned on asyncio debug mode

[cicdw]

Closing this as we've made (most of) these adjustments elsewhere.

[jlowin]

Reopening following offline conversation that the remaining enhancements in this PR may be desirable

[cicdw]

Minor request to reuse an existing attribute, otherwise this LGTM

[cicdw]

we already have an _is_running attribute that you should probably reuse instead

[cicdw]

LGTM - this will be a nice QOL improvement