Add Armenian characters

[silvandiepen]

Hi Sindre,

Hereby a PR, the characters are correct and checked by an Armenian. The only thing is that sometimes in a way the replaced characters get switched around. I can't really figure out why.

example:
When I slugify: Ե ր ե ւ ա ն
It compiles to: re ye v a n, where it should be ye re v a n.

Hope this can be a good addition to the package. So far it works for me with the current replacements as custom replacements, although they arent completely correct, the slugs atleast look "quite ok".

The test in this version passes, but

	t.is(slugify('Ե ր ե ւ ա ն', {lowercase: false, separator: ' '}), 're ye v a n');

Should actually be:

	t.is(slugify('Ե ր ե ւ ա ն', {lowercase: false, separator: ' '}), 'ye re v a n');

Thanks!

[sindresorhus]

The only thing is that sometimes in a way the replaced characters get switched around. I can't really figure out why.

Can you open an issue about that?