Various name fixes, in particular regarding von-prefix (prelast) & lineage (Jr./Sr./I/II/III/IV/V)

[ludopulles]

This pull request contains multiple changes:

• von prefix: A prefix for the last name (van / van de / de la / von / d' but not Van / Van de / O') was sometimes incorrectly seen as part of the last name, which caused the abbreviation (if using e.g. alpha.bst) to be incorrect like [van ] / [de ]. As explained in rationale behind some particle names #107 (comment), by using a 'good' .bst-file in your LaTeX file that sorts on last name not including prefix, alphabetic sorting happens as expected. Even with alpha.bst you have abbreviations like [DvW21] already which is as expected.
• Example, I changed "Lo{\"i}c {van Oldeneel tot Oldenzeel" to "Lo{\"i}c van {Oldeneel tot Oldenzeel}" because without the {}-braces, Bibtex thinks "van Oldeneel tot" is the von-prefix (according to the TameTheBeast manual) which it isn't.
• Lineage was sometimes taken part of the last name when it's not Jr./Sr. but a number, e.g. William E. Skeith III used to be wrong. I forced an encoding of von Last, Jr., First when a lineage appears, to make sure it's parsed correctly.
• Many Belgian names did not include the last name, e.g. I changed "Gilles Van Assche" to "Gilles {Van Assche}".
• Many Spanish/Italian/Portuguese names had a similar issue, e.g. I changed "Alfredo De Santis" to "Alfredo {De Santis}" (similar for "Ivan {De Oliveira Nunes}").
• Changed Maroccan names, e.g. I changed "Youssef El Housni" to "Youssef {El Housni}".
• Many ePrint entries have no spaces, where there should be spaces, e.g. in [EPRINT:Lu23] I changed "Frank Y.C. Lu" to "Frank Y. C. Lu".
• Various name fixes, removing unparsed special characters that appear as "?", e.g. [EPRINT:GaiKiaRus2] has author "Peter Gaži".

I used a script to more or less automate these changes. I can make a PR for these scripts in db_tools and the lib (for changes to Person class) repositories if wanted. I believe that the import script db_import/import.py needs to be revised to do these names good in the future.

Note: this does not update the ePrint labels, also because there was an open pull request #252 . Once this Pull Request is done, I have a fix for all the >6 letter labels and outdated labels, which is here: https://github.com/ludopulles/crypto_db/tree/fix-252
I'll create a PR for that, once/if this is merged.

Closes: #107

[ludopulles]

The scripts I used are found in https://github.com/ludopulles/cryptobib_tools/ . In particular the files:

• fix_von_prefix.py
• fix_lineage.py

Note that to use the "Person" object from mybibtex I modified the cryptobib/lib repo to specify the name output format.