score:3
Proleptic Julian for most dates in ancient history. See here, last paragraph.
There is a third possibility, by the way: For events from the Roman Republic, dates are often given according to the Roman calendar at the time. I recall reading in one modern text that an event happened on the last day of the year and in another that it happened on 29 December--and then I realized they were both right!
As for the founding of Rome, the source of the date is ancient Roman writers. It corresponds to a festival on the ancient Roman calendar. So it's closer to being the calendar of the Roman Republic than Julian or Gregorian (but obviously, that calendar also didn't exist in the 8th century BC). The point is, it's not like we have the date in some other format and decided it likely equals 21 April in Julian. The Romans of ca. 1st century BC called it 21 April (or more accurately, so-and-so many days before the Kalends of May), and we just copied their description.