coverage

We have a problem: if files aren't loaded/required we don't have branch data. If they are added through track_files we give it 0/0 branches which we show as 100% coverage (all possible branches are covered).

That math is "wrong" though here because there are branches but we don't know what they are. We should probably count total branches here as "unkown" and establish that in our "math unive

It would be helpful if the lcov generated reports could generate relative paths. My use case is that I am generating coverage reports from a docker image and then using VS Code Coverage Gutters which supports relative paths to view the coverage. But it doesn't work because the absolute paths on the docker c

Export headers are nice.

undercover should create warnings on entire files that weren't required by specs and hence don't show up in the lcov output.

This can be done by changing logic starting from Undercover::Report#load_and_parse_file

Is your feature request related to a problem? Please describe.
We have a database schema which includes temporary objects - this means objects which are required only for a short time like procedure for upgrades or objects which are still work in progress - and legacy test code. Temporary objects are always prefixed with the username of the author like:
windowsusr_my_table
Test Procedures a

As for our project , we need more detail coverage data about every js file, so can you add the following info to classElement.

        classElement.setAttribute("lines-covered", "" + file.getCodeLinesCoveredCount());
        classElement.setAttribute("lines-valid", "" + file.getCodeLineCount());
        classElement.setAttribute("branchs-covered", "" + file.getBranchesCoveredC