Add support for macrons (vowel length-marks)

[Fuco1]

This would probably need a lot of work over the dictionary, but if we make macrons/lengths supported in the code, the database/dictionary can simply be slowly updated "on the fly".

[mk270]

That's a great idea - there are probably good ways of attacking the lack of macrons in the dictionary, too

[Fuco1]

We've been talking about this in #emacs IRC for years, and there are some programmers and people who would put in some work. I might send some of them here.

Good that you took up the initiative, thanks for that :)

[mk270]

see https://github.com/mk270/whitakers-words/blob/master/plan.txt

[Fuco1]

I was thinking about how to do this and here is my plan:

1. We should make it possible to input text with macrons, but then strip them (simple search&replace) and pass this "fixed" text to later processing.
2. We should make the code loading the database/inflections/stems be able to work with macrons, but strip them before parsing happens. This means that we don't have to touch the parsing logic at all (and it is not even necessary since latin is unambiguous enough without macrons)
3. Make the output respect macrons, so that the user sees it when reviewing the output.

This will have a minor inconvenience such as the user inputting exercitūs which we would also parse as nominative singular (which it isn't). I think this is fine. Later, the forms not matching the input can be filtered in a post-process step.

With these changes in place, the database update can be gradual and everything will work in a backward-compatible way.

Steps 1 and 3 should be relatively simple. Step 2 will also require some changes in the utilities generating the binary files---I don't yet understand the purpose of all of them.