changes to update resources based on predetermined interval

[kumaritanushree]

What this PR does / why we need it:

Kapp will update resources based on predetermined interval set. To set the interval user needs to add annotation kapp.k14s.io/renew-duration to the resource

Which issue(s) this PR fixes:

Fixes #224

Does this PR introduce a user-facing change?

New annotation has been introduced called kapp.k14s.io/renew-duration to set the duration at which user want to renew their resources.

Additional Notes for your reviewer:

Review Checklist:

• Follows the developer guidelines
• Relevant tests are added or updated
• Relevant docs in this repo added or updated
• Relevant carvel.dev docs added or updated in a separate PR and there's
  a link to that PR
• Code is at least as readable and maintainable as it was before this
  change

Additional documentation e.g., Proposal, usage docs, etc.:

Proposal

[praveenrewar]

I will take another look at the implementation in some time.

[100mik]

A bunch of comments!
Main intention being to decouple renewed resource from versioned resources.
Still need to think about what we call things finally.

Kudos for the exhaustive tests, it handles every case I could think of! We should verify it does not add a flake because 2s might be less for slower hosts

[100mik]

How the feature is structured looks great to me!

@kumaritanushree mentioned that today if the user chooses to recreate on update, the last-renewed annotation is not retained. Maybe we should verify if the rebase rule is working.

I do not see a lot of harm in the timer being reset in this case (right away at least). But worth verifying. I am pointing towards the rebase rule as it looks like the feature works otherwise in other case as we check existing resources directly.

Lets also check the tests for flakiness since there's sleeping involved.

TL;DR Next steps:

1. Verify rebase rule
2. Ensure tests are not flaky (a few re runs on the GitHub action worker maybe?)

[praveenrewar]

Apart from the small nits, it looks good to me.

[100mik]

LGTM!
I would prefer not having the unstopped functions, but dedicating a new structure to each one of them feels a bit extra

[100mik]

I think I would prefer it if this function and the next one are either scoped to a struct, maybe we can move stuff that are used by other parts of the code to a single place at a later point. Not a blocker but a nice to have

[sethiyash]

+1

[sethiyash]

LGTM!