Make "Duplicate all" test run faster

[matuskalas]

I see 2 eventual options for making this super-lengthy test faster:

1. Maybe a more efficient, "super-fancy" SPARQL query?

2. Use Python to find duplicates in an indexed collection. For example along the lines of: Use SPARQL to output all labels and concept/etc. IDs; read into Pandas to find duplicates; return the output in a Caséologue style.

It is really a "show-slower" in all work on the GitHub Actions and Caséologue itself, waiting 10-80 minutes after each push... 🙁

In either case, advise from @albangaignard please 😊

@maanst is motivated to work on this 🙌🏽

Note: Please also look at the unit test for "duplicate all" in test_caseologue. It is disabled for this reason currently.

[matuskalas]