Haskell source not general enough

Currently, the Haskell source assumes that the slip direction, the rupture propagation direction and one side of the rectangular fault are parallel to each other. This should be relaxed such that at least the rupture propagation direction and slip direction can be independent.

Thanks to Luis Rivera to point this out.
