Bug in QL MDP for New Graph

Seems as though there is a bug in the QL model. I say this because, it still tends to look at all the arcs with computation cost set to a small positive number. In the Exact MDP implementation it selects a subset of the arcs. I will investigate further.

I also fixed a bug where it was not correctly counting up the arc lengths correctly. I tested it with zero computational cost, and it gets the same answer as the plan all strategy.