Issues with current implementation of `rpi_compression`

[matiasandina]

• [x] animal_id has to be passed to compress_videos
• [x] the files we care about are not all .mp4 files. They have patterns and we have to exploit that
  • [x] session_id has to be extracted from the files
• [x] style: 'sub-{}_ses-{}.{}'.format(animal_id, session_id, file_extension) -> f"'sub-{animal_id}_ses-{session_id}.{file_extension}'
• [ ] elif file_extension == '.csv': is not a thorough condition. The files we are looking for have a specific format (see camera.py output)
• [ ] no else to handle patterns that do not match (including .csv and .mp4 that should not be compressed)