Releasing the DoReCo CLDF dataset

Install requirements:
```
pip install cldfviz[cartopy]
```
Adapt the URLs for languages metadata and citations in cmd_download to the ones linked from https://doreco.huma-num.fr/languages
Adapt/remove the fixes in cldfbench_doreco.py if the corresponding issues in the "raw" data have been fixed.
Download the data running
```
cldfbench download cldfbench_doreco.py
```
agreeing to the defaults whenever prompted.

Re-create the CLDF dataset running

cldfbench makecldf cldfbench_doreco.py --glottolog-version v4.8 --with-cldfreadme --with-zenodo
cldfbench readme cldfbench_doreco.py

Make sure the data is valid running
```
pytest
```

Make sure data can be loaded into SQLite

rm -f doreco.sqlite
cldf createdb cldf/Generic-metadata.json doreco.sqlite

Commit all changes, tag the release, push code and tags.
Create a release on GitHub and make sure it is picked up by Zenodo.