You raise a very important and, I think, complicated point here. "I maintain what I said before, and repeat my point, which you did not answer: If Craig defends genocide, ALL the more reason to debate him."

I think that something like this is true, but this itself is not true.

If Craig defends genocide--and it is hard to call it anything else given the passages Dawkins quotes--then his arguments need to be refuted. This is critically important in an enlightenment-based society: you simply cannot afford to allow calls for widespread violence to go unanswered. You might end up deciding that the violence is necessary (to end a war of aggression, perhaps) or forgivable (from a less civilized, more desperate time), but you at least need to talk about it.

So, Craig's ideas should absolutely be examined.

But this does not tell us either who should do the examination or what venue should be used.

People do not always believe things because they have engaged in sound reasoning. There is a wealth of literature from psychology attesting to this, demonstrating all sorts of hilarious or disturbing influences on people's judgments, revealed preferences, and stated beliefs. In particular, people tend to be motivated by tribal allegiance, respect for authority, and a variety of other factors.

So if you believe someone is wrong in a harmful way, it's also legitimate to consider whether in debunking their arguments you provoke non-reasoned adherence to their perspective. Rarely if ever does this mean that a bad idea should go unchallenged, but you may wish to employ some tactical thinking in how it is accomplished.

Debates are typically viewed as contests between people of roughly equal standing (e.g. the Pope does not, generally, debate new members of a random university's Atheist Club), about matters where reasonable people can disagree (e.g. we don't generally have debates about whether bats are birds), and where an extensive back-and-forth is required (e.g. if we disagree about the distance between Earth and Mars, we look it up (or measure it)--we don't debate it).

That means that by entering a debate, you presumptively grant all of these things. But maybe, instead, you want to not grant all of these things because you don't want to elevate a purveyor of bad ideas beyond what is necessary to debunk the ideas. As Dawkins says: "I have consistently refused, in the spirit, if not the letter, of a famous retort by the then president of the Royal Society: "That would look great on your CV, not so good on mine"."

There is the further danger that any debate would not be on what Dawkins calls out as Craig's most reprehensible ideas, thereby potentially allowing Craig an opportunity to put on a good showing on some front and thereby (in a logically invalid but psychologically unavoidable way) make his reprehensible ideas seem more palatable.

So, that goes for debate about genocide itself and about anything else Craig cares to debate. If Dawkins relished debate, sure. If not, fine. Someone else can instead.

I nominate Stephen Woodford.

Except Craig could very well say, "That would look great on your CV, not so great on mine."