Feature: Added script for generating and saving a local leaderboard

[Pringled]

Description

Added a script to create and save a leaderboard locally for a given benchmark as discussed in #2011.

The script retrieves results, parses them to a Pandas DF, prints them, and (optionally) saves them to a CSV. It uses the same functionality used for generating the live leaderboard. This makes it much easier to generate results locally and check aggregate scores before submitting them to MTEB.

Usage

The script can be used as follows:
python scripts/make_leaderboard --benchmark "MTEB(eng)" --save-path "leaderboard_results"

Optionally, a custom results repo can be used (default is https://github.com/embeddings-benchmark/results), and specific models can be requested, for example:
python scripts/make_leaderboard --benchmark "MTEB(eng)" --results_repo "https://github.com/Pringled/mteb-results" --models "potion-base-8M" --save-path "results"

Code Quality

• Code Formatted: Format the code using make lint to maintain consistent style.