I totally agree that this is an unacceptable failure on the part of the engineering teams.
To one of your points, and it remains unclear to me, those rigs might actuate only the chassis without the engine running. McLaren seems to have a few more details on their site.
http://www.mclaren.com/formula1/car/simulated-secs-how-f1-teams-find-laptime-1147741/
I had read that article from James the day it was posted, and it was what I thought of when I read your post. I interpreted his article to mean that the constructors are effectively running the cars on rigs at the factory, which is one of the things that allows them to amass so much test mileage once they are at the track.
The things I have been reading now suggest harmonics may be vibrating the electronics apart, which is not a good scenario to be in. Have experienced similar in the MR2 I race in. When a race engine gets up over 100 hours, the bearing clearances cause more vibration, we start getting weird failures like distributors and ECUs.
To be clear, I really wasn't trying to discredit your post, I was simply trying to add more information to the discussion.