Coverity defects should be detected and allowed to block integrations into Qt, Mobility, other products.

The most sensible way to do this is probably to transform the list of Coverity defects into QTestLib-compatible logs so they can be stored alongside existing test results. This would avoid having to implement any special Coverity-specific results comparison or blacklist logic. The generated test logs could hopefully point to the defect in Coverity web interface for full details.

One likely challenge is the Coverity runtime. If we are not able to reach ~4 hour Coverity runtime or faster, it will not be workable in our current CI setups.