Issue filter: allow users to further process a filter using a python script

There will always be cases where filters are bit too dumb to fit a certain situation. Thanks to @bretdavi, we can now execute arbitrary python code in a restricted separate process. This could be used to allow users to add the following scripts which would only get executed if all the filter's requirement are first fulfilled.

# Inputs:
#    test: Name of the test
#    testsuite: Name of the testsuite
#    status: Result of the test
#    machine: Name of the machine that run the test
#    machine_tags: Array of tag names associated with the machine that run the test
#    stdout: Standard output of the test
#    stderr: Error output of the test
#    dmesg: Kernel logs outputted during the execution of the test
#    run_log: Output of the test suite's run
# 
# Available modules: re
#
# Outputs: returning True will tell CI Bug Log that this filter matches on this test result.

return 'gem' is in test and 'FUNKY ERROR' is in dmesg