- Revamped archive detection mechanism to use virtual host name rather ...

than a request variable.

- Split off separate headers and footers for each repository. ...

- Made "default" header and footer generic.
- Fixed bug in mysql class that grouped eprints with the same ID from
  different archives.
- Fixed non-XHTML compliant image elements.
- Changed eprint title links so that they point to the appropriate stats
  page in the correct archive. (Especially relevant to the "default" stats).
- For the "default" stats, included a note (+ link) indicating which repository
  each eprint is stored in.