Dataset licensing considerations

[hlibbabii]

This issue summarizes all the concerns/tasks related to licensing of datasets used by or produced by BOHR.

Current situation

Currently, dataset/desc field in bohr.json is used to mention the license of the dataset along with attribution and the description of the dataset itself. There is also dataset/author field.

Ideas what needs to be done/improved:

• separate field license for each dataset in bohr.json
• Field to specify where the dataset has been downloaded from. This might overlap though with the corresponding DVC file if the dataset is "imported".
• Since dataset descriptions are getting bigger, should we consider having a separate e.g. datasets.json for that. Alternatively, should we have a file per dataset? Can we use dataset stage files (.dvc files) for that - .dvc file should have a description field?