Add ability to run all TCK unit tests within Hudson, showing success and failure rate

Currently our use of the JCR TCK unit tests (again, not the official TCK but a subset of what's run within the TCK) is handled within a JUnit test suite in the "dna-jcr" project, and as we develop new features we're uncommenting tests as we pass them. This does not give us any information about how we're progressing. We want to run the tests (perhaps nightly) and make these stats available to let people know how we're doing and proceeding towards Level 1 and Level 2 compatibility.

Comments, thoughts, and approaches welcome.