Feat: Evaluate missing languages

[Samoed]

Checklist

• Run tests locally to make sure nothing is broken using make test.
• Run the formatter to format the code using make lint.

Continue ideas from #1525

If a missing split is present in the existing results, return its results without running the evaluation.

[KennethEnevoldsen]

Looks good here only a few minor things

[Samoed]

I have a small issue. When I run mteb run -m intfloat/multilingual-e5-small -t Banking77Classification --overwrite with existing results, it doesn’t do anything and I don't understand why.

INFO:mteb.cli:Running with parameters: Namespace(model='intfloat/multilingual-e5-small', task_types=None, categories=None, tasks=['Banking77Classification'], languages=None, benchmarks=None, device=None, output_folder='results', verbosity=2, co2_tracker=False, eval_splits=None, model_revision=None, batch_size=None, overwrite=True, save_predictions=False, func=<function run at 0x7fc7bbf42440>)
INFO:mteb.models.sentence_transformer_wrapper:Model prompts will be overwritten with {'query': 'query: ', 'passage': 'passage: '}
INFO:mteb.evaluation.MTEB:

## Evaluating 1 tasks:
─────────────────────────────────────────────────────────────────────── Selected tasks  ────────────────────────────────────────────────────────────────────────
Classification
    - Banking77Classification, s2s


INFO:mteb.evaluation.MTEB:

********************** Evaluating Banking77Classification **********************

[isaac-chung]

I have a small issue. When I run mteb run -m intfloat/multilingual-e5-small -t Banking77Classification --overwrite with existing results, it doesn’t do anything and I don't understand why.

This runs fine for me in your branch.

[Samoed]

Hm, intresting

[isaac-chung]

Looks good. Just a few suggestions :)