🚀 go-pugleaf

I happen to receive e-mails whose charset indicated in the headers does not
match the actual contents, for instance the mail contains latin-1 characters
while it is flagged as us-ascii, or windows-125?, while it is flagged as
latin-1. When I reply to such broken email, and insert latin-1 characters in
the reply, and eventually send the mail, it gets encoded in iso-2022-jp (I
guess vm did not convert properly into MULE the incoming mail, so that one
gets a weird mixture of charsets in the reply, and iso-2022-jp ends up to be
chosen to encode the whole mess).

This is not a bug in vm (it is a bug in the MUA of the sender), but it is
nevertheless a very annoying feature :-). I don't have much spare time these
days, so I have cowardly decided to post that feature-report instead of
investigating the issue deeper, I hope you will not blame me for that :-).

If you can think of a solution, I would be grateful.

PS : this feature was there in 7.07 that I used until today, I checked 7.11
before posting this, and the feature is still there.