organization of the generation of the metadata files

[SylvainTakerkart]

I'm looking at the generation of the metadata files (and I'm using this issue to take notes of my reflexions...)

For a dataset, the main function generate_all_metadata_files is called for each instance of the BEP032Data class, i.e, in the original phylosophy of the design of the class, for each session...

This raises several questions / problems at different levels:

• for patchclamp datasets, since the session concept is not applicable, overall, we have a problem... this is to be discussed in the context of our soon-to-come "global refactoring" brainstorming
• more generally, even for extra-cellular ephys datasets, I don't see why the generation of the dataset_description.json file is called for each session... is there a reason for this? (if no, this might enter the "global refactoring" brainstorming)