Add `--debug-worker` flag to make debugging workers easier

[dgollahon]

It is difficult to debug workers because they capture all output. This is relevant for hooks as well as for general mutant development. I have often resorted to doing something like File.write('something.txt', some_value.inspect) as a very rough workaround.

[hiSandog]

A useful next step might be to separate the repro into input, state/cache update, and final output for --debug-worker. That would make it clearer whether “Add --debug-worker flag to make debugging workers easier” is failing during parsing/configuration, during internal state updates, or only at the user-visible result.

[mbj]

I'm working on mutant 1.0 that moves the entire engine to Rust (via prism and a new high speed rust based unparser). That has this feature already. Problem its not a feature parity yet when it comes to mutation operators and some compatibility checks.

If super pressing I can add the feature to the "ruby mutant", raise your voice if needed.