✨ Using date breaks/minor breaks/labels in time scales.

[teunbrand]

This PR aims to fix #4335.

Briefly, it pulls scale_{x/y}_time() into the datetime_scale()-fold of functions.
I'm at a slight 'Chesterton's fence'-risk because I don't know why scale_{x/y}_time() was excluded in the first place, but I'll take the risk.

[thomasp85]

I don't like the name date_breaks for this because a date has a fixed position in time whereas the time scale is relative. But for symmetry with the other scales it is probably worth this minor gripe

[teunbrand]

I also don't think the names are great, but I'm not a big fan of the date break mechanism to begin with.