Well, emphasising this wrinkle in CI is rather unrevealing, IMO. The US already has rules about soluble and insoluble fiber for its labels.
Specifically, that insoluble fiber is 0 calories, and we want lots of it in our diet, despite its being a category of 'carb'. FDA food energy content calculations are not ignoring this fact, it's baked into their calorie estimates. As is bioavailability in general. There is still error bars to the calorie estimates, though, for various other reasons. For example, carrots of the same varietal still vary quite a bit in composition depending on when and where they were grown. Some have more starch, some more water, by mass, just as an example.
Yay for anecdotes. Is it the "low carb" you are claiming "works" independently of CICO?
This is why I like Hall's approach, and am a bit relieved I couldn't find any significant flaws with his experimental protocol. Specifically, the experiment was double blinded, which addresses what I call the 'fanboi' problem with so many diet studies. Diets are not immune from biased perception, selective memory, and other expectation effects. "The doctor told me this will work, so I'll stick with it," is especially relevant where diets' relative effectiveness is heavily dependent on adherence over time.
I'm reminded of an analogy from when I was a kid. My dentist had bought into applied kinesiology. He wanted to prove to me that it worked, so he asked me to resist his downward push on my arm, then tapped a probe on one of my teeth and tole me my arm was now weakened, and he pushed down on my arm again. I expect he'd shown this to other people and found their arm had weakened. But mine didn't. Undaunted, he doubled down and just short of jumped onto my arm to push it down. "See! Your arm got weaker." (No Doc, you were pushing harder to protect your belief, I appreciate it may not be a conscious decision). I consider this to be one of my pre-Skepticism skeptical experiences.
For Hall's experiment, neither the patients nor the doctors knew the composition of the food, and we can see that absent expectations and hype, composition does not appear to impact appetite or weight management. But adjusting calories does. No surprise, but it's nice to see an experiment that addresses a lot of potential weaknesses in other studies so directly.