Make number parsing locale aware

[rhdunn]

At the moment tts/context_analysis.cpp only handles numbers of the form nnnnn. That is, it does not handle numbers of the form:

`n,nnn,nnn` -- e.g. US numbers
`n nnn nnn` -- e.g. Swedish numbers
`n.nnn.nnn` -- e.g. German numbers

This should really be handled in tts/text_reader.cpp which should be passed a "digit separator" character. Doing it this way allows tts/text_reader.cpp to produce a single number token.

This should be done based on the associated document language.

--- Want to back this issue? **[Post a bounty on it!](https://www.bountysource.com/issues/1026783-make-number-parsing-locale-aware?utm_campaign=plugin&utm_content=tracker%2F254961&utm_medium=issues&utm_source=github)** We accept bounties via [Bountysource](https://www.bountysource.com/?utm_campaign=plugin&utm_content=tracker%2F254961&utm_medium=issues&utm_source=github).