Let me just play the role of the devil's advocate here for a bit.
Date, time, timezones, etc. and their implementations have what we call Essential Complexity. It's important to understand the complexity involved in any case and we can't make it much easier.
That said, of course it's possible to introduce a data type that only stores a date, but the next feature request will be to do calculations between regular datetime objects and these dates, or to convert between them, etc. Then you're back to square one, except it's now a more complex model.
Another argument is that there are of course cases where you don't need the time component, but you'll still have to understand localization, when to use it and how to convert it for all the other datetime attributes in your system.
These are just a few of the reasons why we've chosen not to do this yet, and of course it would take a lot of time (no pun intended)