Put the subcmd name in the validate_output error message [#1539]

[genehack]

Description of proposed changes

This puts the full sub-command name into the error message shown when the output records have inconsistent field names.

Related issue(s)

#1539

Checklist

• Automated checks pass
• Check if you need to add a changelog message
• Check if you need to add tests
• Check if you need to update docs

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 69.90%. Comparing base (81badd2) to head (58db430).
Report is 176 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1540      +/-   ##
==========================================
- Coverage   69.90%   69.90%   -0.01%     
==========================================
  Files          75       75              
  Lines        7882     7888       +6     
  Branches     1933     1933              
==========================================
+ Hits         5510     5514       +4     
- Misses       2086     2088       +2     
  Partials      286      286              

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

[joverlee521]

My comment in #1518 came back to bite real quick 😅

there's no good way to check for the output error. Hopefully we just never see it!

[genehack]

My comment in #1518 came back to bite real quick 😅

there's no good way to check for the output error. Hopefully we just never see it!

I spent like, 30 seconds thinking about mocking something up to cause the error so I could write a test, then settled for manually validating it on the broken version of parse-genback-location...

[joverlee521]

Once this is merged, I think good to release along with #1531 as Augur 25.1.1

(Probably wait till Monday so we don't do a Friday release 🙈)

[genehack]

I spent like, 30 seconds thinking about mocking something up to cause the error so I could write a test, then settled for manually validating it on the broken version of parse-genback-location...

In the harsh light of morning, I realized, while functional tests would be a PITA, I could totally trivially unit test the validate_records() method, so I added that.