Hold a sheet of paper in front of your mouth; blow over it (on the top side only). The paper should rise, the pressure falls thus the paper rises. And in this instance how to you define the ‘bottom’ surface.
Similar effect blowing through a paper tube - a paper bag open at both ends. The sides move in due to air moving through it and reducing pressure - reduced with respect to ambient.
So for either the paper sheet or tube, if the ambient pressure has not changed then the surface over which the airflows faster must contribute all of the lift - reduced pressure with respect to ambient. cf JT # 8.
Re the question, the proportions of lift should to be considered and stated relative to a datum pressure (and AoA, wing section, etc).
An example would be to normalise the pressure plots in post #2 so that the lower surface is datum (zero), then all of the lift is from the top surface because the top surface ‘arrows’ have increased due to the added lower surface ‘arrows’ being reversed.
Over to the tech-science views who might wish explain otherwise.
Oh, then there is the rotating tube, Coandã effect, in still air …
And a wedge shape wing section in supersonic flow where one surface at the leading edge is parallel with the airflow …
And my simple view of the lower wing surface is so the top surface can be fixed on it to provide lift, and structural strength enabling mounting on the fuselage.