Once the club gets to a reasonable angular velocity it will develop an  inertial resistance to changes in the orientation of the plane on which is it is moving.
HK , in the glossary, mentions the gyroscopic action to be the cause for this resistance to any attempt to change its plane.  I don't agree.

With gyroscopes one deals with fast spinning symmetrrical device such as disks.  Its action is based on conservation of angular momentum.  In a golf swing we don't have a  fast spinning symmetrical device.  Instead we have arms and club starting form zero speed, continuously accelerating and building up angular momentum instead of conserving it.

A less esoteric but more practical way to look at this issue is to take into consideration the centrifugal force and how it might play a role.  As soon as the club moves away from the plane a restoring torque is developed.  I made a first order of magnitude calculation fort he bottom part of the downswing and the reuslt is shown in Fig1.

The equilibrium angle between the shaft and the sweet spot plane is about  -.5 deg which is due to the COM offset of the head, otherwise it would have been zero deg.  This equilibrium situation corresponds to the situation as shown in Fig2.  

The restoring force has a quadratic relation with the angular velocity,  hence swinging twice as fast results in a fourfold increase in the restoring torque.

mandrin