You only need matter with negative mass if you want to construct one by building a machine. Some models for closed timelike curves that could occur in nature (e.g. the Tipler cylinder with infinite length) do not need negative energy.
> the likelihood of FTL being possible is extremely small, for all reasonable purposes indistinguishable from zero on current scientific knowledge, and the existence of entanglement does not change this one bit
Okay, that I agree with, but this unlikelihood is completely orthogonal to (pre-relativity) QM's compatibility with FTL travel/signaling. That was what I disagreed with.
> You only need matter with negative mass if you want to construct one by building a machine.
As I'm sure you're aware, things do move faster than light and/or backwards in time on the microscopic level. But that's not what most people care about when they talk about this stuff. They want to know if you can build a time machine. And the answer to that (on current knowledge) is a definitive "no".
You're right that GR doesn't rule out closed timelike curves, but to actually construct one you need to have matter with negative mass. So you can interpret this in two ways, as a prediction that FTL is possible, or a prediction that negative-mass matter does not exist.
Nonetheless, since you clearly understand physics, I'll restate my claim more precisely: the likelihood of FTL being possible is extremely small, for all reasonable purposes indistinguishable from zero on current scientific knowledge, and the existence of entanglement does not change this one bit. It would take something like the discovery of exotic matter with negative mass to move the needle on FTL.