#3163 Don't include the DwcOccurrence id column in DwC exports

[kleintom]

No description provided.

[mjy]

@kleintom This is by design. The id is not persistent. To recover records we need to reference the occurrenceID.

[codecov-commenter]

Codecov Report

Attention: 2 lines in your changes are missing coverage. Please review.

Comparison is base (a428145) 84.76% compared to head (1a7b598) 84.76%.
Report is 23 commits behind head on development.

❗ Current head 1a7b598 differs from pull request most recent head 9e92649. Consider uploading reports for the commit 9e92649 to get more accurate results

Files	Patch %	Lines
app/helpers/collection_objects_helper.rb	0.00%	1 Missing ⚠️
...pp/models/dataset_record/darwin_core/occurrence.rb	0.00%	1 Missing ⚠️

❗ Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

@@             Coverage Diff              @@
##           development    #3789   +/-   ##
============================================
  Coverage        84.76%   84.76%           
============================================
  Files             2100     2100           
  Lines            77394    77417   +23     
============================================
+ Hits             65600    65623   +23     
  Misses           11794    11794           

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

[mjy]

@kleintom This may be introducing failures: https://github.com/SpeciesFileGroup/taxonworks/actions/runs/7587358876/job/20667528644?pr=3789#step:21:74

[kleintom]

That worked out great actually, it

• caught a bug I had introduced: DwcOccurrences.meta_fields was opening the newly created csv file, yanking out the headers into an array, and then shifting off the first one, which used to be id. Now the shift is unnecessary. That fixed all but one of the failing tests (those were all checking meta_fields).
• The other fixed failing test was checking the header fields in an exported dwc csv file - the check now correctly checks that id is not included.
  (I squashed the commits so there wouldn't be failing tests on any of them.)