corry binary is relatively minimal, which is great. Hence, I had the idea that it might be nice to have a python version. This could serve as a starting point for people trying to code up more complicated analysis workflows with multiple corry runs in a single script without having to rely on shell scripts. Ultimately, it would be nice to externalize the step of reading the config file object into the python code, so that could be manipulated on-the-fly from the python side, but let's take one step at a time.
Please provide feedback on these changes. I have tried to make them as uninvasive as possible, but it might be that I missed some important details here and there. The most crucial aspect is that I generate the python bindings with ROOT/cling. I was hesitant to do so at first because I am not sure how deep corry intends it relationship with ROOT to be, but seeing how ROOT is already a de-facto dependency for the object streamers, I feel less bad about also using it for python bindings.