chore(data): Remove data extracts (-50) ...

And ignore big .gz files

chore(data): Remove BOM from doi-vol3.txt

chore(data): CRLF to LF (for doi-vol3.txt)

fix(data): Remove Doi header

chore(data): Add volume 3's DOIs

feat(detect-cnrs): Add the CNRS detecting ...

oui or non in ws.cnrs

chore(data): Update sample data ...

Using last .ini scripts

feat(harvest-from-doi): Add the script

chore(corpus): Move corpus to the root directory ...

Because it's hard to put a big file in a sub-directory (especially via
webdav).

fix(enrich-pascal): Increase timeout to 1 minute

feat(enrich-pascal): Add script ...

And pascal.sh which is a curl test, because enrich-pascal.ini does not
work.

refactor(make): Use make without parameters to produce final json

feat(extract-field): Concat addresses & reprint_addresses into aff

feat(enrich-teeft): Use abstract AND title

chore: Add a Makefile ...

And regenerate corpora

fix: Chaque notice a un abstract, et c'est une chaîne

fix(enrich-teeft.ini): Use patched ezs packages

chore(teeft): Ask for 10 keywords

feat: First commit