Document the subprocess API

[jvkersch]

          @mikedbjones Both sound great, thanks for picking this up!

1. I have indeed snuck a few pieces of the subprocess API into main, without impacting the existing functionality. If you could open a PR to update the README to describe this, that would be a very big help. You could add a new section, and you could describe the new functionality as experimental. To be clear, I want to remove the existing implementation (at which point the new implementation will no longer be experimental) but I don't know when that will happen.
2. For the symmetrization, yes, please feel free to contribute this as a PR. Would you be able to add an example of a nonsymmetric travelling salesman problem, and a short description of how the algorithm addresses this?

Originally posted by @jvkersch in #28 (comment)