Store Shakespeare texts in repository rather than downloading and caching them

[rgrp]

Why do this?

1. We started off downloading the data on demand. However we also needed to clean it up (remove Gutenberg cruft etc) and transform it. Once you start need doing some of this by hand you cannot just 'download' and clean automatically and hence we need to store 'our' copy somewhere permanently.

2. (Relatedly) We need to put text in svn so we can track changes.

3. By bundling with svn the user gets all the material on install and Open Shakespeare 'works out of the box' -- without any issues about users internet connections etc etc.

[rgrp]

Done in r153 and r154.