Add rddlsim output parser

[thomaskeller79]

If we want to compare planning performance to other planners (e.g., SOGBOFA), we can either write an individual parser for every other planner or we can use the rddlsim output as a generic way to collect the per-round reward and average reward.