[ODE] Bug in limot with dJOINT_REVERSE?

[Matthew D. Hancher]

Heyo.  It appears that the limot code does not respect 
dJOINT_REVERSE.  At the very least, this seems to cause 
reversed joints to have inverted velocity.  It may also 
affect joint limits and such; I haven't thought about 
it very much yet.

It seems to me that the best way to fix this would be to 
check joint->flags within dxJointLimitMotor::addLimot().
Does this seem reasonable?

mdh

Matt Hancher
NASA Ames Research Center
mdh at email.arc.nasa.gov