Release: 0.1.0

• Exports the roster in two formats: by appointment
  (roster_by_appointment.csv, which was roster.csv) and by judge
  (roster_by_judge.csv, made by roster_by_judge), and run returns them
  as "roster_by_appointment" and "roster_by_judge".
• Keeps each judge's professional career in the roster as career, one
  position per line, beside the fields made from it.
• Writes each judge's name in twelve ways for matching (name_first_middle_last
  and the rest), with the FJC's bracketed letters removed or replaced by a
  period.
• Adds the FJC's sequence of each appointment to the roster.
• Adds optional Judicial Common Space scores (add_jcs, --jcs), computed
  by judicial-common-space (0.1.0 or later, from PyPI), and the justices'
  Martin-Quinn scores (add_martin_quinn, martin_quinn_terms,
  --martin-quinn).
• The tests save the roster of their invented judges in both formats in
  tests/output, to be looked at.
• Adds a citation file and a BibTeX citation in the README.