[FEATURE] Decouple data for keyword/primitive/stories etc. and make importable

[saxoncameron]

It's probably inevitable that we cannot keep some of the data in our current kanji db due to potential issues with licensing or copyright - e.g. data from keywords and primitive names etc. from RTK, stories Kanji Koohi, etc. This would only pertain to relevant data like keyword, primitive name, stories, et. al. and not the other character metadata and linked words etc.

Short of acquiring permission or some other agreement, a logical solution would be to totally decouple these datasets from the extension to allow for the community to make contributions and import/load datasets according to their preferences. E.g. there might be a WaniKani one, an RTK one, a Kanji Koohi one, a Kanji Damage one, etc. Whether those are official or unofficial wouldn't be our concern. Importing/stacking multiple datasets would need to be supported to. And of course they'd all need to adhere to some defined.

I think the addon should come with, by default, an crowd-sourced and fully open-sourced community created and curated dataset. For the separate ideas I have around how that might work, I have created a separate issue #113.

[saxoncameron]

This could also be a way to implement the Japanese keywords dataset for advanced learners #85.

[sihue]

In the meantime, would an import/export function with kanji,keyword,story(,userdata) csv files be an option? This would be a little more convenient for backups and shared spreadsheets than manipulating user_files/user.db every time. Import functions like overwrite, only import if empty and concatenate would also be neat.

[saxoncameron]

Not a bad idea, and perhaps a quicker in-the-meantime style solution. Unfortunately, this is still quite low on the priority list, as ~1.3.0 is a loooong way out unless demand for this bumps it up for us. Some bug fixes to do as part of 1.0.x too