fix: duplicate score sets appearing in GET results

[bencap]

This pull request improves the handling of superseded score sets to prevent duplicate results and enhances the clarity and structure of related tests. The main change ensures that when multiple unpublished score sets supersede the same published score set, only one instance of the published score set is returned to non-owners. Several test cases are updated for consistency and readability, and a new test is added to verify the deduplication logic.

Score set deduplication logic:

• Updated get_experiment_score_sets in experiments.py to deduplicate score sets by their IDs, ensuring that the same visible ancestor score set is not returned multiple times when traversing superseded chains.

Testing improvements:

• Added a new test test_non_owner_searches_multiple_unpublished_superseding_score_sets_no_duplicates to verify that non-owners see only a single instance of a published score set even if multiple unpublished score sets supersede it.
• Improved formatting and structure in several tests (test_update_experiment_keywords, test_update_experiment_keywords_case_insensitive) for better readability and maintainability. [1] [2]
• Minor formatting fixes in assertion statements for error messages in tests to improve clarity. [1] [2] [3] [4]