Update body and wake FoR multiplication

[kccwing]

silly mistake really - position in Gframe of multibody is offset by cga_0 not cga(t)

[ben-l-p]

Apologies for taking so long to getting around to looking at this PR - this is an issue I was facing, so thank you for the fix. Should the time step you index not be the current one self.data.structure.timestep_info[self.ts].cga(), rather than the initial time stepself.data.structure.timestep_info[0].cga()?

[kccwing]

Thanks @ben-l-p, no worries! The trouble here is that the multibody solver, unlike the standard structural solver, stores in timestep_info['pos'] the displacements in the A frame, rather than in G - this is a patch to correct for that. Ideally this should have been done in the solver structure I agree - but the code uses the position too often so this is a "cleaner" patch.

[kccwing]

ah sorry, I think I didn't answer your question - so basically the position is correct - it is just forgetting the initial FoR change.

[ben-l-p]

Okay, that makes sense. Will approve it now, however will merge once we add a patch for the Mayavi issues, which I will investigate today