There were more Romans abroad in the first half of the first century than you might think. In this "Roman" category I also include those who had been "Italians" before the Social War (90-88 BCE) and instantly became Roman citizens under the lex Iulia. The Italian merchant class had enclaves in every corner of the empire and beyond, though we only get scattered references to them. For instance, when Mithridates invaded the Roman territory of Bithynia in Spring of 88 BCE, he was persuaded to perpetrate a radical action: to kill all the "Romans" there. Our sources for this are pretty lousy (Appian and Plutarch), but Appian says 80,000 "Romans" (this includes "Italians") were killed in a single stroke. Even if he is doubling his figures, it still seems to be a significant number. Of course not all of these "Romans" were fighting men, but you get the idea.

Similarly: when Pompey fled to Greece during the Civil War, the backbone of the forces he raised were technically Roman citizens. We don't know the percentage, and we do hear about distinctions between Spanish "auxiliary" infantry (vs Roman legions), but it isn't exactly clear how that worked. I suspect every legion had a core of Romans, especially the centurions, and were filled in as best as could be managed. A while ago I wrote on this topic in this subreddit, and said this about Pompey's forces in the Civil War:

Lentulus Crus, technically a consul, had raised two legions in Asia. We know these were Romans because we know that Romans of Jewish descent were exempted in this conscription effort thanks to his decree, preserved by Josephus (Josephus, Antiquitates Judaicae, 14.228ff). Metellus Scipio was to bring two additional legions from Syria, and Pompey's own son went to Egypt to round up what he could find. These legions might have, and probably were, under-strength, but they formed the citizen core of Pompey's army. By the time of Pharsalus, we hear of legions from Cilicia on the right wing, supported by auxiliary infantry from Spain. We know he also sought and obtained massive numbers of auxiliaries from the various remaining client kingdoms in the East, and he even sent overtures to Parthia asking for support (Dio 42.2.5). Cicero was certainly disgusted by the notion of Pompey's not-so-citizen legions bearing down on Italia: " I—whom some called the preserver of this city, some its parent—I to bring against it armies of the Getae, Armenians, and Colchians! I to inflict famine on my fellow citizens, devastation upon Italy!" (Cic. ad Att. 9.10) (link)

Even back in 58, Caesar was bending the rules a bit as far as legionary recruitment. When the Helvetii made their move, Caesar tells us that he immediately began raising legions in the extant Gallic provinces. We know that he also recruited among the "Transpadani," a group mixed Gallo-Italians living in the Po valley. They were not technically Roman citizens, though some communities there had various levels of Roman privileges, and we know that Caesar promised those recruits full citizenship after their service.

As to the number of Roman colonies: the short answer is "a lot, but not nearly as many as would come later." Before the 50s BCE, the biggest were in Spain (Italica, Corduba, Valencia, Segovia), France (Narbonne), and of course in Italia (Ariminum, Placentia, Salernum, Bononia, Aquileia, etc), and very very few in the East. That means that most of the legions raised in the east by Pompey in 49 were not from colonies, but from these elusive groups of "Romans" present for other reasons. The really crazy colonial period in Roman history comes after the second Civil War, the one between Octavian and Antonius, when Augustus settled tens of thousands (possibly over 100k) of troops in new colonies in places all over the Med, from Spain to North Africa to the Near East.