You haven't being attention. CO2 is not the problem for diesels; NOx is.
Now that's out of the way, it seems to me that there's a reasonably simple way to catch both types of cheating.
1) Borrow a car from a dealer chosen at random. Stuff a gas analyser up its tail pipe and drive it at least 62 miles/100 km. with the analyser logging its readings along with gradient and speed
2) put that car and at least one more through the statutory rolling road test cycle.
Now look at the results. Examination of (1) should spot the Chrysler-Fiat style of cheat: if there's a step change after the test cycle length you've got them. Comparing the two traces should spot ECUs using the VW type of cheat because the road test will show higher emission levels at all stages of the drive than the rolling road test does. This approach may well pick up hybrids of the two as well as new and unforeseen ways of cheating.