Move batching from Task to LLM, fix vLLM.generate and add DISTILABEL_LOG_LEVEL

[alvarobartt]

Description

This PR moves the batching from the Task to the LLM so that the LLM handles the batches in the best way possible rather than via a simple for-loop, since there are some LLM engines that have mechanisms to handle batches in a more efficient way. Also the prepare_input abstract method has been removed from LLM and is no longer needed, unless a specific LLM implementation requires it.

Also this PR fixes vLLM.generate and stops propagating the unsolicited inputs through the Pipeline, so that only the ones solicited via the property are kept.

Besides that, this PR also adds the DISTILABEL_LOG_LEVEL environment variable to control the log level of distilabel, which defaults to INFO.

Example

Find a full example at https://huggingface.co/datasets/alvarobartt/instruction-dataset-mistral-7b-instruct-v0.2/blob/main/example.py

[plaguss]

🚀