The dark arts, as I understand the phrase, are techniques which can be used to persuade people about a position even if the position is false. Therefore logic isn't a dark art, because it's pretty hard to persuade someone about a falsity using logic. Ad hominem is a dark art, because it signals that the opponent has low status and people are biased to align their opinions with high-status peers, irrespectively of truth or falsity of the said opinions. Charitable debating style alleviates the status motivations and I don't see what other biases it can exploit to promote a falsity.

But since we should rely on empirical evidence, let's test it. The idea of the test is to select a false proposition (not trivially false, as ("2+2=17"), but also not subtly false ("there are ontologically fundamental qualia") or too technical ("there is an error in the Wiles' proof of Fermat's last theorem") and try to convince people about it on two comparable forums. On the first, use the techniques suspected of darkness, on the second, refrain from using them. Compare the numbers of convinced opponents.