Fix incorrect returned values when getting ascendants_and_descendants

[mgeplf]

• When ascendants_and_descendants are chosen, and there are entities in the precise within_brain_region_brain_region_id, they would be retrieved twice - once from ascendants, and again from the descendants
• pagination would limit the returned objects, but could include the duplicates, which were then removed by the unique in db.execute(data_query).scalars().unique()
• this resulted in the number of objects returned being lower than what was requested in the page_size

[mgeplf]

See https://github.com/openbraininstitute/prod-build-ion-channel-model/issues/50#issuecomment-3710516911 for the bug; hat-tip to @bilalesi for the report.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

Flag	Coverage Δ
pytest	97.60% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
app/filters/brain_region.py	100.00% <ø> (ø)

🚀 New features to boost your workflow:

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

[bilalesi]

I was skeptical about union_all, thanks @mgeplf

[mgeplf]

This also should fix:
#474
As noted by @GianlucaFicarelli

[eleftherioszisis]

This also should fix: #474 As noted by @GianlucaFicarelli

Would it be difficult to also include a small test that covers the #747 issue just to maintain our sanity with potential regressions in the future?

[mgeplf]

Would it be difficult to also include a small test that covers the #747 issue just to maintain our sanity with potential regressions in the future?

That's what the page_size is doing here:
https://github.com/openbraininstitute/entitycore/pull/498/changes#diff-da4e1cbf774ac6257d994abd0e710ad69fadf968cc4e3480cd9a75055408ea49R389

Before the fix, it would have returned 1 entry, instead of 2.

[eleftherioszisis]

lgtm