Integrate translations more closely in W3C web site

[dontcallmedom]

The previous translations system was maintained as a set of static RDF files where the data on translations and translators was semi-automatically kept up to date.

While that system served us well, it became harder and harder to find the resources to maintain it, and its operations as a set of static RDF files deprived us of a closer integration into other W3C systems.

Some of these integration points would be:

• linking directly from a spec to the translated document in the user's language (if available)
• linking all the documents someone contributed to translate from that person's profile page on W3C
• managing language negotiation for documents where appropriate

[dontcallmedom]

The new approach would:

• have the database tracking translations fully integrated and maintained as part of the W3C systems used to manage our specifications; these translations would be exposed vi the W3C API, allowing their re-use in more contexts (e.g. third-party maintained lists of translations on a given topic)
• re-use our existing github/W3C integration points to keep track of the identity of translators and facilitate their recognition

The tighter integration would enable e.g. to bring translation in profile pages (assuming they get made available, which hasn't been decided yet), as well as provide automated links from TR documents themselves through a JavaScript library-powered insert.