And on the same page Pauli also said that if the mirror has any motion perpendicular to the plane of the mirror, doppler shifting would occur. Which in itself implies that if the observer has any motion perpendicular to the mirror in the mirrors co-moving frame, then doppler shifting will also occur.
Tachs fundamental mistake, well, one of them anyway, is he proves that the angle between the instanaeous velocity of a particle on the rim of the wheel, and the tangential velocity is zero in the co-moving frame of the axle of the wheel (Duh! I could have told him that!) and then argues because it is zero there, it must be zero in all frames. But what he has failed to account for is that if the wheel is moving in an observers rest frame, then the observer must be moving in mirrors co-moving refrence frame. And so if the motion of the mirror is the salient consideration in the observers rest frame, then the motion of the observer is what we must consider in the mirrors co-moving frame, not the motion of the mirror.
But it's worse than that, and even more fundamental than that, because for the reasons I demonstrated, but Tach dismissed out of hand, even a co-moving observer would measure doppler shift that is dependent on on the angle from their line of sight, because on one side of the line of sight, if the observer was to measure the distance between them and a point on the surface of the wheel, they would find it move towards them, and on the other side of their line of sight, it would be moving away from them.