[MIEB] feat: add jina-clip-v2 to MIEB

[bwanglzu]

NOTE, this is a draft PR since the model hasn't been released yet.

Add our latest model jina-clip-v2 to MIEB, similar as jina-clip-v1, aims at train on text retrieval and image-caption retrieval tasks, while it features several new characteristics:

1. It supports 89 languages, become a multilingual multi-modal embedding model.
2. it supports MRL (truncate dimensions).
3. It has basic capability of understanding documents (visually rich document retrieval), reaches similar performance as So400m/Bipali on ViDoRE and our internal benchmark, the image tower could be used as a visual backbone for VLLM training.

Compared to jina-clip-v1, it started to use a prompt when encoding queries to improve text retrieval performance, so i added an additional parameter task (could be discussed) similar as jina-embeddings-v3.

Checklist

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

Adding datasets checklist

Reason for dataset addition: ...

• I have run the following models on the task (adding the results to the pr). These can be run using the mteb -m {model_name} -t {task_name} command.
  • sentence-transformers/paraphrase-multilingual-MiniLM-L12-v2
  • intfloat/multilingual-e5-small
• I have checked that the performance is neither trivial (both models gain close to perfect scores) nor random (both models gain close to random scores).
• If the dataset is too big (e.g. >2048 examples), considering using self.stratified_subsampling() under dataset_transform()
• I have filled out the metadata object in the dataset file (find documentation on it here).
• Run tests locally to make sure nothing is broken using make test.
• Run the formatter to format the code using make lint.

Adding a model checklist

• I have filled out the ModelMeta object to the extent possible
• I have ensured that my model can be loaded using
  • mteb.get_model(model_name, revision) and
  • mteb.get_model_meta(model_name, revision)
• I have tested the implementation works on a representative set of tasks.