Reading reconciliation/recongraph from .csv

[aremath]

Pull request #173 changes the output format of reconcile to .csv. I think it would be good if there was a way to read the .csv back to create a corresponding reconciliation or recongraph data structure. This would allow someone to (for example) reconcile many files using a shell script that calls the CLI, then read them in to the gui individually to look at the graphs or other outputs, rather than having to load all three files separately each time. The reason this is not implemented is that the "best roots" are not saved with the .csv, and it seems like there should also be a refactor to obviate the need for "best roots" by creating a fake "start" node in the reconciliation graph that saves the best roots as its children.