CFR/CFT - GSoC 2014 Unicode Collation Algorithm implementation

[Pedro Giffuni]

Hello;

Last year Dmitry Selyutin implemented collation as his Google Summer
of Code 2014 and I was tricked into mentoring it ;).

My understanding is that this is a very desired feature for FreeBSD 11
and while Dmitry did a good job and was generally successful[1], the code
needs heavy reviewing and testing from people that are likely to depend
on it (that means you).

I have put together his code for Code Review:

https://reviews.freebsd.org/D2736

For those that prefer a patch, It is also available as a diff, here:

https://people.freebsd.org/~pfg/patches/gsoc2014-ghostmand-collation.diff.gz

The code builds and generally should work, but I haven't done enough with it
to claim it is a complete solution or that it should be committed. Also, 
if people
want to replace the use of bdb with something else like NetBSD's constant
db, then it needs to be done, but at least I thought I would make ithis 
visible
as a starting point.

Best regards,

Pedro.

[1]
https://lists.freebsd.org/pipermail/freebsd-i18n/2014-August/000437.html