The official term is "cogging", and it depends on more than you think.
Basically, the motor wants to relax into a position where the magnets are as close as possible to as much iron as possible. Pull the motor away from this position, and it wants to pull back. Pull it far enough, and it goes "over the hill" and it wants to advance to the next cog instead of the one that it had been on.
All motors that have an iron armature that rotate with respect to the magnets exhibit at least some cogging torque. There are, in fact, high-zoot motors available that have windings that are molded "in the air" and have no iron attached -- search on "coreless motor"; they're pretty much de rigueur for servo motors, both RC and industrial.
This cogging torque depends on the strength of the magnets, and on the width of the gaps in the armature, and on how much gap there is between armature and magnet, and on the width of the gaps between the magnets, and on whether the gaps in the armature are staggered or not, and probably on a few other things that I don't know of or haven't remembered. Motor manufacturers can significantly reduce the cogging torque without affecting the overall motor torque much by constructing the armature so that the gaps are a bit skewed to the axis of rotation -- that way, the magnets "see" the gap at different angles of rotation along the length of the armature, which has the effect of making the gap "look" softer to the magnets, and you have less cogging torque. This is probably what eFlight does.
I don't think you need to be hugely concerned about this. Cogging torque is an issue, because it's always there and may be part of what makes motor bells resonate at just the wrong frequency. But except for certain motors, I don't believe that its a huge issue.