Make requirement coverage depend on test fix version

[John Milasch]

As pointed out in this question, the requirement status is not calculated based on the test fix version. This creates problems when the requirement has a slightly different implementation in 2 different versions. When this happens the steps in the test for one version might have to be different than the steps in the test for another version. 

If we create 2 tests for the requirement, and we execute for one of the versions only, the requirement coverage will be listed as TODO and this is not desirable.

I'm requesting that an option is added to the administration page that the calculation of the requirement coverage is also dependent on the test's fix version.