Maximum Edge Orientation improvements

[agitter]

#34 added initial support for the Maximum Edge Orientation pathway reconstruction algorithm. After adding support for partially directed input and output graphs (#28), this algorithm will need to be updated. We also need to document the assumptions that this wrapper makes and the required arguments (#26).

It is also possible to load the MEO paths to rank edges. The SDREM software does this as an extension of the MEO code, but it is not directly supported by MEO yet.

As the MEO maintainer, I should also update its readme to advertise that it is now available (and easy to use) via SPRAS.

[annaritz]

Quick note here - I merged main into the precision-recall branch before I continue working on it, and the Test Docker Wrappers failed during reconstruct with MEO. I suspect this may be due to some input file formatting issues since it's running on a netpath pathway. I haven't looked closely at it yet. See Pull Request #36 for the full output.

[agitter]

I'm noting that MEO may have problems with pairs of directed edges (u,v) and (v,u). That will need to be explored further once SPRAS supports partially directed input networks.

[agitter]

MEO also crashed with undirected self-edges when I was testing a version of the EGFR dataset. I removed those edges from our example input data those in #120.