Restricting quaternion sign interferes with odeint

[translunar]

I narrowed down the problem to translunar@a2da759.

When using odeint with qdot for quaternion propagation, sometimes odeint gives errors that look like this:

Solving ODE
 lsoda--  warning..internal t (=r1) and h (=r2) are
       such that in the machine, t + h = t on the next step  
       (h = step size). solver will continue anyway
      in above,  r1 =  0.1198705359199D+02   r2 =  0.8575714648946D-15
 lsoda--  warning..internal t (=r1) and h (=r2) are
       such that in the machine, t + h = t on the next step  
       (h = step size). solver will continue anyway
      in above,  r1 =  0.1198705359199D+02   r2 =  0.5846098197682D-15
 lsoda--  warning..internal t (=r1) and h (=r2) are
       such that in the machine, t + h = t on the next step  
       (h = step size). solver will continue anyway
      in above,  r1 =  0.1198705359199D+02   r2 =  0.8353495261073D-15
 lsoda--  at current t (=r1), mxstep (=i1) steps   
       taken on this call before reaching tout     
      in above message,  i1 =       500
      in above message,  r1 =  0.1198705359199D+02
/home/mohawkjohn/.local/lib/python2.7/site-packages/scipy/integrate/odepack.py:218: ODEintWarning: Excess work done on this call (perhaps wrong Dfun type). Run with full_output = 1 to get quantitative information.
  warnings.warn(warning_msg, ODEintWarning)

I suspect that one of the propagated quaternions is developing a negative scalar component, pyquat is detecting that and negating the entire quaternion, and that creates a discontinuity that odeint observes.

I'm not sure how to fix the problem at the moment (or if it should be fixed, since it isn't technically a pyquat bug), but I'm listing it here so other people don't get caught by the same problem.

Copied over from https://github.com/autumnsault/pyquat/issues/8