Spanwise flow is simply not the cause of tip vortices. It is a largely incidental local flow phenomenon. If you don't like the earlier thought experiments, imagine a lifting, finite span wing with a huge tip plate. Huge enough to push spanwise flow as far from the surface of the wing as you want. Will there be a tip vortex? If there isn't, tell me how the air provides the reaction force to the wing.
The air is able to provide the reaction force in the form of a pressure differential between the top and bottom of the wing. Any -additional force- provided by deflecting flow downward is an inefficiency of a wing.
An inefficiency you say? YES! Because the other side of the vortex pushes air UP. It takes energy to do this. That energy is taken away from the pure pressure differential of the idealized 2D wing. This is why a 3D wing always has less lift than a 2D wing (at least within linear theory).
This is the exact opposite of how a jet propulsion system works. In a jet propulsion system, force is created by momentum transfer. Any additional force provided by a pressure differential between the exit plane and the atmosphere is an inefficiency.