I think I disagree with your definition of the words "often" and "massive". Some cyclists do have muscular legs (relative to the average person), but it seems like the "biggest" legs on cyclists are on the sprint cyclists - the guys who pedal at much higher intensities for shorter distances (more comparable to lifting in that low-to-moderate rep range vs doing dozens and dozens and dozens of reps).
13,064 Pull-ups in 5 Months from 2011.
I don't think GVT is the best comparison for what you're discussing. If anything, crazy-high rep/high frequency bodyweight training is comparable to the old "one day arm cure" where you do a few sets for bis and tris every few hours over the course of a whole day.
Again, sorry but I'm not sure we're talking about the same thing throughout. Low weight, high rep sets are pretty much only good for building endurance, not muscle. But once we increase total volume (doing 6x50 vs. 1x50, for example) then it's a different animal and you start, eventually, pushing the stimulus for growth, however inefficient and roundabout it may be.
Go back and look at how Waterbury set up that pull-up experiment. It was lots of easy/not to failure sets in the 7-15ish range done throughout the day to accumulate a high training volume. He wasn't doing one set of 30 in the morning, another set of 40 in the afternoon, and a set of 50 at night.
Long story short (too late), higher rep work can build muscle if you're doing enough total volume, but it's an odd method to choose because it requires the use of significantly submaximal weights. Doing lots of sets for lots and lots of reps with a light weight won't build any strength worth noting, so... what's the point?