Fix rejection reasons from first review (2020-02-26)

TODO: maybe open sub-issues for things that take longer time. It might be best to try to address issues upstream.

From the rejection mail:

I can't find any indication that the project is licensed under LGPL-2.1+ as opposed to just LGPL-2.1.

tests/lang/eval-okay-fromjson.nix contains text from an RFC whose license does not permit derivative works (it is not explicitly marked as a code component in the RFC).

doc/manual/style.css says "Copied from http://bakefile.sourceforge.net/, which appears, licensed under the GNU GPL."

scripts/install-multi-user.sh appears to have been sourced from a number of webpages; it is not clear that it has been released under a DFSG-compatible license by all its actual authors, so probably needs to be filtered out.

It is not clear that the mk/ directory is the same license as the rest of the project, as no license is given in the GitHub repo referred to as the canonical source. Please ask upstream about the intended license.

doc/manual/images/callouts looks to be vendored from docbook, and the license of docbook requires copyright notices and license text to be copied into d/copyright.

+----------------------+ | Other comments | +----------------------+

As previously mentioned, the upstream author's name has been used in first entry in d/changelog, instead of the uploader's name.

README says "This product includes software developed by the OpenSSL Project for use in the OpenSSL Toolkit." but I didn't spot any openssl code; please ask upstream about whether that notice should still be there.