Option for Setting Gauss-Seidel Value Iteration

[etotheipluspi]

Currently, the solver only supports Gauss-Siedel value iteration. We should have an optional flag to do value iteration without the Gauss-Siedel method if the user wishes. There are a few ways to do this. The two most straightforward ways:

1. Have two different solver types with their respective solve functions

2. Have a flag that looks something like gauss_siedel::Bool as a field in the current ValueIterationSolver type without adding another type or solve() function.

Can this be done more elegantly? @zsunberg if I recall you were dealing with something similar to this in MCTS, any input here?

[mykelk]

I'm okay with either.

[zsunberg]

My gut feeling is that the flag is adequate since most of the functionality and options are the same. Actually I didn't think too much about this for MCTS. It may have been better to do dpw with a flag.