Support different file types for I/O

[finozzifa]

When working on the DEA energy storage parser, I noticed that the technologies.json file, when all rows from the raw excel data sheet are included, has a size of about 10 MB and contains almost 200k lines.

[euronion]

More general questions that follow from this:

• How do we setup data distribution with this / other packages?
• Do we distribute data with the package and in the repo? Alternatively: Do we ask/require users to parse data everytime they want to use it?
• Do we stick with json? Or switch by default to something else? (csv, compressed json, parquet, sqlite, ...)?

[euronion]

For now, could explore zipped json files, as they are simple to implement with our current setup.