crashfrog,
Well, then let me ask you the same question I asked Buz: why?
For exactly the same reason that anything gets classified in the Linnaean system, they possess enough characters that as a body are common to the class reptilia. In exactly the same way that lizards, crocs, plesiousaurs, pterosaurs, & turtles get classified as reptiles. Remember, we aren't talking cladistics & synapomorphies that all "reptiles" must possess in order to be considered reptilian. Take undifferentiated teeth, for example. Turtles have beaks. Toothed whales have undifferentiated teeth, too. But
overall, taken in conjunction with scaly skin, egg laying etc is is an indicator of "reptilian-ness". Same goes for warm/cold bloodedness. Certainly it would make the headlines, but a warm blooded lizard found today, that possessed all the other key indicators would still be classified a reptile.
I've checked the more recent sources on my bookshelf, & they all agree Dinosaurs are reptiles.
You can of course hypothesise that cold/warm-bloodedness is such a key indicator that it warrants a class level move, but you need to convince everyone else of that. Science is a consensus activity.
This is an analogy, since cladistically reptiles don't exist at all. But I doubt putting a single character state "warm-blood" into a cladogram that has hundreds of character states to consider, is going to see dinosaurs shifted anywhere else than where they are already on the generally accepted cladogram (represented all amniotes clades).
There is no reason warm-bloodedness rules out a reptilian classification in and of itself.
Mark
There are 10 kinds of people in this world; those that understand binary, & those that don't