For very large numbers, no efficient, non-quantum integer factorization algorithm is known.
Note: While common consensus is that no efficient algorithm exists, it has not been proven to be the case. To prove such a thing would be equivalent to proving that P = NP -- in other words it would require solving one of the major unsolved problems in computer science. For more on how NP and complexity-theoritic reductions relate to zk-snarks see this excellent post by Chrisitian Reitwiessner.
The documentation at https://docs.iden3.io/#/guides/circom-and-snarkjs?id=_23-toy-example includes the factoring decision problem as an example circuit, and says:
In actuality, the factoring decision problem is widely believed not to be NP-hard, which would mean that an efficient factoring algorithm is not sufficient to establish P = NP. See https://en.wikipedia.org/wiki/Integer_factorization#Difficulty_and_complexity for a summary of what is known.
There is also a misspelling of "complexity-theoretic".