Add Fuzzy Versions, to compare bytes <> socrata

[alexrichey]

This modifies a few things:

• Adds fuzzy version comparisons in the BYTES <> Socrata script so that we can compare version formats like '2025-01-01' with '2025Q1', or "Oct 2025" with "2025-10-01" to see what's potentially out of date. E.g.

• Modifies the output dataframe to make results more actionable. Namely, it sorts:
  1. Products with ANY out of date datasets, where there exists a Socrata version
  2. Products with ANY out of date datasets, but no Socrata version (these may be out of date... or they may just be TODO on the Socrata side, and so have no version)
  3. The rest

This PR is mostly slopped. I took a shot at integrating this with our existing dcpy.utils.versions code, but came away thinking that will be a bigger endeavor (and probably refactor of that module).

[codecov]

Codecov Report

❌ Patch coverage is 53.03030% with 31 lines in your changes missing coverage. Please review.
✅ Project coverage is 75.65%. Comparing base (751229e) to head (65884dc).
⚠️ Report is 14 commits behind head on main.

Files with missing lines	Patch %	Lines
dcpy/lifecycle/scripts/version_compare.py	53.03%	29 Missing and 2 partials ⚠️

Additional details and impacted files

Files with missing lines	Coverage Δ
dcpy/lifecycle/scripts/version_compare.py	41.74% <53.03%> (+41.74%)	⬆️

... and 4 files with indirect coverage changes

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[fvankrieken]

Just thinking about trying to trim down a little...

Could we declare original and normalized as typed class attributes? I don't see how this first if clause would ever be met.

Then, since we're not returning "exact match" or the like, just true either way, should we just compared the normalized versions? We've already computed them and they should be identical if the originals were identical. Meaning we could skip the second if clause as well.

[alexrichey]

Yeah, I'm with you. Made it terse. For the first case (if you're referring to if not self.original or not other.original) it'd be when it resolves to None == None which want to be False.

[fvankrieken]

I got bad news bucko

[fvankrieken]

Wow my reading comprehension is terrible today sorry

[fvankrieken]

can this be q|Q? I feel like we see both

[fvankrieken]

Sorry! We've lowered at this point

[alexrichey]

indeed - though latest commit adds an explicit test case for that (there was something close before, but not quite) and changes the comment

[damonmcc]

sorry why aren't these probably equal?

edit: oh direction matters! if something is Q1 it could be any of 3 different months so we'd rather not say these're equalish