Add schematron processing and validation to the build

[joeytakeda]

Stemming from #73 and part of #206: we need to add schematron process and validation to the build. I think we should do what we've done before and use the XSLT's provided by schematron to extract from RelaxNG (if /where necessary per #207), convert the schematron to XSLT, and then store the XSLT in the repository as part of the schema build.

For the actual staticSearch build, we should then have another process where we can pass a document (or set of documents) to the produced schematron XSLT for validation (I think both @martindholmes and I have similar bits of code for this already).