Having read many such articles over the past 20 years, I more and more believe that the scientific evidence is mostly supported by revenue. Trying to make susceptible buyers buy lenses they do not need.
My better half has been using a Nikon D800 for about 5 years now. She uses mostly the Sigma Art primes. 35mm f1,4 and 135mm f1,8. Both lenses are tested with high resolving power.
I just bought a D800 two days ago, and will mostly use lenses like the Nikon 55mm f1,2, 105mm f2,5 and 35mm f2 Nikkor-O. I decided to buy a 36 Mp camera after using the lenses on Kristinas D800 and seeing how they perform there.
From those personal experiences, the least sharp lenses are the 55mm f1,2 (wide open) and the 300mm f2,8 Ai-S. All other lenses do not only appear sharper, but, like the 4 element 135mm f2,8 Nikkor-Q.C actually are perfectly sharp when zoomed to 1:1, which is closer examination then most viewers will perform on the resulting images.
Optical constraints is another term for "We want your money". So far I have not seen evidence of lenses for the Nikon F mount not being perfectly sharp on cameras with an F mount. Maybe I'm just lucky though, and buy all the good samples out there