Fix Exception to remove result from duplicates

[eileenmcnaughton]

Overview

This updates, and tests, the Exception.create function (via the api) to remove duplicates from the
prevnext_cache. This feels like a better fix than the cartwheels the prevnext cache code does to avoid
retrieving cached duplicates.

Before

Marking a non-duplicate does not remove from cached duplicates table (although it is excluded on retrieve I have some doubts as to whether that works reliably when the pair has been flipped)

After

Cached contacts removed

Technical Details

This feels like a more reliable approach to me. I didn;'t go down the path of proving whether there was an issue with existing code when flipped as I think this is worth doing anyway

Comments

[civibot]

(Standard links)

• If this is your first pull-request for CiviCRM, please browse CONTRIBUTING.md for information about the development and testing processes.
• If you are reviewing this pull-request, you may wish to consult the test sites and the Review Standards (long template, short template).

[eileenmcnaughton]

test this please

[eileenmcnaughton]

@pfigel I'm not sure if you have a specific interest in my efforts to make dedupe functionality api-accessible but it you do - here is another...

[pfigel]

@eileenmcnaughton yep, we'll probably need to work on our deduping process at some point this year, so anything making that more accessible to extensions is greatly appreciated!

• General standards
  • (r-explain) PASS
  • (r-user) PASS: Provides a more stable implementation of the same behaviour from the user's point of view, no user impact.
  • (r-doc) PASS
  • (r-run) PASS: Confirmed exceptions are removed from civicrm_prevnext_cache after API call.
• Developer standards
  • (r-tech) PASS: Improves implementation
  • (r-code) PASS
  • (r-maint) PASS
  • (r-test) PASS

[eileenmcnaughton]

@pfigel we'd definitely love to collaborate with you on it - this is the extension I've been working on - https://github.com/eileenmcnaughton/org.wikimedia.dedupetools - trying to make the manual process more efficient since we already have a bunch auto-deduped