Rework "update freshness" to not error with library -> ingest

[fvankrieken]

We are going to have a failing run this weekend without these changes.

#1097 added specific logic on how to handle an ingest run when we've already archived that version of a dataset. However, it didn't account for when the previous archived "vintage" of that version had come from library. Given that we're in the middle of this migration, and there are some slight data changes occuring (that would cause the validation from #1097 to error), I decided to clean up the code a little and simply "pass" - no overwrite, no updating timestamps - when the existing version came from library

Commits are quite atomic. I moved around some testing code hence the several commits, it should make commit-by-commit review easier. I thought pulling out the logic of "validating" (@sf-dcp 's favorite word) the new dataset versus existing versions to a function which makes no changes, and then based on the enum returned run decides what to do.

Integration Tests

Running in my dev bucket, first run a job via library by using a branch of mine without latest main, so opendata job still uses library. This should get us latest versions of datasets, archived by library. I cancelled it as to save runtime. There are also the DOT ones which went private and failed.

Then, run a job on main with my dev bucket. See failures (except for the ones that weren't actually archived in the previous step). There are also some internal server errors happening? 500 errors. It seems like we might be getting rate limited for real. Anyways, see this job here to see "successful" failure

Then, run a job on this branch with my dev bucket. Getting one bizarre s3 error that I'm not going to try to troubleshoot this second

[codecov]

Codecov Report

Attention: Patch coverage is 90.24390% with 4 lines in your changes missing coverage. Please review.

Project coverage is 70.07%. Comparing base (a041b06) to head (c9ddc2e).

Files with missing lines	Patch %	Lines
dcpy/lifecycle/ingest/run.py	66.66%	2 Missing and 2 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1339      +/-   ##
==========================================
+ Coverage   70.00%   70.07%   +0.06%     
==========================================
  Files         114      115       +1     
  Lines        6121     6145      +24     
  Branches      702      706       +4     
==========================================
+ Hits         4285     4306      +21     
- Misses       1690     1692       +2     
- Partials      146      147       +1     

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