tree f2dfc99525473ea5af1c02b20521fa9680ae5265
parent 81777013dfa582053489efea874e84379f1f0805
author Matthias Andreas Benkard <code@mail.matthias.benkard.de> 1785065546 +0200
committer Matthias Andreas Benkard <code@mail.matthias.benkard.de> 1785065546 +0200
gpgsig -----BEGIN PGP SIGNATURE-----
 
 iJEEABYKADkWIQRE3QzuUV5Uegq6sdJPLjDfCicO+wUCamXwShsUgAAAAAAEAA5t
 YW51MiwyLjUrMS4xMiwwLDMACgkQTy4w3wonDvsmZQEAofJ4SK4rfPJhQwo5VVmC
 UycD9ylPqii3Fk2BPhtUU3cBAI1gOAxZLgPfR2+ESZHq1hor7yLaNYD08jPeis4S
 Mp4O
 =LGGt
 -----END PGP SIGNATURE-----

Bring the documentation up to the current state

The README still described state law as Bavaria-only, six Länder later.
The sample-data overview grew a section on how to obtain the pre-
amendment stems per portal — which of the three routes works, and that
the juris state portals are a dead end — and a short prioritised "where
to continue", so the next session does not have to re-derive it.

Corrected while writing it: the regression figures are not all pinned
the same way. UWG, AGG and ProdHaftG pin "nothing left manual", BayJG
pins 154 commands and names its five renumbering residues, and the state
cases pin full application. Saying "UWG 19/0, GEG 66/53, …" would have
described observations, not what the tests actually assert.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Change-Id: Ie247e9626e789e628d537bede735923d5b67d14b