Threat each newline as separate word in diffWordsWithSpace

[Mingun]

This fixes #180 & #211

[Mingun]

@kpdecker any feedback?

[Mingun]

@kpdecker, can we make progress with this? Better late than never

[kpdecker]

Sorry for the delay on this, Have very few cycles for this project these days.

[Mingun]

Nice to see that you don't forgot the project

[ExplodingCabbage]

IMO the behaviour this produces for diffWords is broken as a result of the extra tokens. Here's a diff using the code before this change...

> wd.diffWords("foo bar", "foo\n\nbar")
[ { value: 'foo\n\nbar', count: 3 } ]

... and here's one with the code after this change:

> wd.diffWords("foo bar", "foo\n\nbar")
[
  { count: 2, value: 'foo\n' },
  { count: 1, added: true, removed: undefined, value: '\n' },
  { count: 1, value: 'bar' }
]

We're now only sort-of-ignoring whitespace in diffWords.

(In fairness, we were already doing a bad job of ignoring whitespace in diffWords; if you changed the whitespace around some punctuation it would be treated as a change, before.)