IanCal and blaze33 both gave good answers, but just to clarify one thing: the radiation we see coming from regions where there are black holes is of the first type: radiation emitted by objects like gas clouds that are falling into the holes, before those objects cross the event horizon. If we leave out quantum effects like Hawking radiation (see below), it's impossible for light, or any kind of radiation, or indeed anything at all, to escape from inside the event horizon of a black hole.
Hawking radiation is a quantum effect, which nobody has ever observed; the reasons for thinking that it exists are purely theoretical.
Isn't the nonexistence of large black holes from cosmic ray collisions proof of black hole evaporation? Or have we not been able to measure that / have reason to doubt micro-black-holes are created by cosmic ray collisions in the first place?
I would say we don't have accurate enough measurements or an accurate enough theoretical understanding to know how many micro-black-holes we should expect to see from cosmic ray collisions, on the assumption that none of them ever evaporate, or to be able to measure how many there actually are, so as to be able to compare the two numbers to see if there's a significant difference. In principle this would certainly be a good experimental test for the existence of black hole evaporation; I just don't think it's a test we can make with any confidence now or in the near future.
Hawking radiation is a quantum effect, which nobody has ever observed; the reasons for thinking that it exists are purely theoretical.