Ben, you are close and I think I know you are saying the right thing but you are mis-using the word turbulence. The "roughness" as we called it and as Vadim showed in his nice photos is to add energy to the boundary layer of the flow to keep it from becoming turbulent. When flows turn, say from the plenum to the runners and into the port energy is lost. Just like on a wing/airfoil as angle of attack increases, etc. in an enclosed volume (say a pipe which closely mimmics that of the port) the boundary layer remaining attached to the wall is what will determine how the flow develops. We want a nice fully developed laminar flow, but thats in a perfect isentropic world where we neglect small perturbations. We hope and assume the flow to be laminar and as its energy (created by differences in pressures and velocity-think Bernoulli's Principle) varies it can become transitional an eventually turbulent. Ingesting turbulent air is not good as the fuel wont mix properly by any means and the combustion process will be less than ideal from the turbulence and poor mixing.
The boundary layer of a laminar flow is very sensitive to shear stresses due to viscous effects or other perturbations on the surface which will cause it to separate and transition to turbulence...whereas transitional/turbulent flow boundary layers are robust (less sensitive).
The roughness we speak about needs to be very little, in fact not rough by our standard means of thinking just not mirror like finishes. As the flow enters the port various transitions are occuring which consume the flows energy and thus the "roughness" for lack of better words creates a skin drag or viscous force that will renergize the BL and keep it attached to the surface giving the flow a better chance of staying as close to laminar as possible in said conditions.
In a laminar flow, the air particles are almost uniformly distributed throughout the field which is what makes for good mixing potential wheres a turbulent flow is exactly what it sounds like, chaos lol non-uniformity more particles on one side than the other, since the fuel is denser than the air, (we all know that) its gonna follow the path of least resistance whereas in a laminar flow all of the paths are "the same" so to speak.
the closest thing I could think of that does this same job is a vortex generator, except this process does like you said, introduce turbulence in order to add energy. BUt VGs are ususally used in flows with reynolds numbers much much greater than that of a head port. such as flow over a large wing or an entire vehicle. In a pipe flow its common to use vanes to straighten flow but obviously impractical in a port so viscous effects is all we have to manage it.
and yes...this conversation went to shit due to the 50 topics. I mentioned the turbine grinding process was a poor mans EH and then 50 different things were mention only to arrive at the conclusion that a smooth surface finish on a turbine housing is good and the methods of grinding on a turbine housing and leaving a rough finish are not good.
The horse has been beaten well after it was dead
Cheerio!