As the slit gets smaller the percentage of photons that are bouncing off of the walls of the slit as compared to the percentage that goes through directly raises.
True, but geometric rays of light bouncing off the walls of the slit won't broaden the beam. Similarly, the barrel of a machine gun doesn't broaden the stream of bullets coming out.
Only Heisenberg does (note, however, this is a purely classical/maxwellian use of Heisenberg).
Right, but it is interesting that if the camera were adjusting it's iris due to the decrease in light, we would see the same thing without Heisenberg. The spread would be there in both cases, but would be overwhelmed in the wide-open case by all of the particles of light making it straight through. As the light intensity lowered and the iris widened to compensate, the effects of the photons bouncing off of the walls of the slit would begin to be more visible.
