Automate representation test overwrite

[bbrockbernd]

This PR fixes two problems when working on trace representation.

• It provides a script that runs all tests on all relevant jdks on all modes (trace debugger and non trace debugger)
• It auto generates txt output files for new tests, only for jdks and modes where the outputs differ (to prevent duplication)
• Since it generates the txt files necessary, we only provide a expectedOutputFilePrefix. The result of this is that a lot of test files changed.
• Since most test files changed (in line numbers) the output txts changed as well.
• Additionally, since it chooses when a seperate txt file for a test is needed, some txts are gone and others have been added.

How it works:
The test outputs of the run with jdk 8 and non trace mode, are considered as the base output and will therefore always (over)write the txt. If a test is run with a higher jdk and/or in trace mode. We will look if a txt for that combo exists and if not we will look at lower jdks (and at the non-trace side). We will then compare the output to the found file. If it differs we (over)write the file with current jdk and mode.