That's a good question. I've just spent five minutes thinking about it and came up with "no", but I'm not really sure. The difference is that the proof in Curry's paradoxon uses logical equivalence (denoted by the "<=>" sign, or "iff") which is inherently circular all by itself: in step n of the proof, I can put in any statement from some step m with 0 < m < n and start again, making it circular. In mathematics, a circular proof usually consists only of implications, i.e. I have a number of theorems T(1), T(2), ..., T(n) and in order to prove all of them, I show that T(1) implies T(2), T(2)->T(3), ..., T(n-1)->T(n) and finally show that a single T(i) for 0<i<n+1 holds true, which implies that all the other theorems are also true.
Another argument in favor of a negative answer is the fact that a proof in propositional logic usually consists of expressions implying other expressions, while circular reasoning consists of theorems implying other theorems, so there's sort of a type mismatch.
Additionally, circular reasoning isn't bad at all. It's a common technique often used in practive, and if we disallow it in propositional logic, we'd also have to disallow it in mathematics.
Greets,
Philip