I don't disagree with the second part. But the ought is explicit in THN, like 'n4r9 says below. I can think of two (definitely not the only two) plausible interpretations:
1. Reason is necessarily the slave of the passions. If so, there is no sensible way in which it ought be, since normative claims require that a thing could be otherwise. If Hume truly means that, then his "ought" is probably more of an admonishment against trying (and inevitably failing) to usurp the passions. This is a different sense, but I think it's close enough in kind to justify the interpretation.
2. Reason't isn't necessarily the slave of the passions. This makes the "ought" clause more intelligible from my position, but isn't really supported by Hume's account of moral action (i.e., how we find ourselves driven to it).
> This is a factual, not a normative claim.
It's normative in the context of the "admonishment" interpretation above. But yeah, not from #2.
> Reason is, and ought only to be the slave of the passions...
However, given the context, I suspect that the use of the word "ought" here has a slightly different meaning than we tend to infer these days.
I don't think you will find Hume actually arguing this. People read this into him because they glide so easily from is to ought.
> or that it must be so
This is a factual, not a normative claim.