I believe that example is not steady state so can't apply to a long trip, and unlike the Leaf examples, they weren't measuring purely the power demand from the heating/AC systems. That means there are other non-heating related factors that can play into the efficiency numbers. Basically the comparison is a lot more useful if you just compare power in kW and don't conflate miles into it. This is because heating demand varies with time, not directly with miles.
The problem with your calculation is you were essentially assuming that there is no consumption by the heat pump, which is far from the case. What matters is the difference.
Right, I acknowledged both that it wouldn't be steady state, and that the heat pump wouldn't use zero energy in my post that you quoted.
So I agree it's the delta that matters. I suspect that delta would be large enough to be useful. Now as to if the implementation cost makes that an attractive option, I don't know.. but given the relatively easy process of adding heat pump functionality to an existing A/C system, along with the success other low-cost EV's have had with implementing them, I suspect it might be.