Weightloss improves various metabolic biomarkers, including markers of insulin sensitivity, and so the fact that people who lose weight following a particular intervention (low-carb or whatever else) see such improvements is unsurprising. That's why looking at subjects who are already lean is much more informative: it gives a better sense of the metabolic effects of the dietary intervention itself, not confounded by weightloss.
> Without that it sounds like the study contradicts a huge body of evidence which it doesn’t.
What huge body of evidence? All of the people drastically restricting their carbohydrate consumption so they can cheat on their fasting BG and HbA1C tests? Give those people an Oral Glucose Tolerance Test and measure their insulin response in real time, and the vast majority of them would fail, especially the ones on high (saturated) fat diets: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6602127/
> Our finding of a strong relationship between VO2max and relatively unsaturated IMCL pools in controls and athletes points to a role of exercise in decreasing the amount of saturated fat within the IMCL store. The association of insulin resistance with the accumulation of saturated IMCLs, even within a healthy control population, could suggest an early involvement in its pathogenesis and provide a reason why combined exercise and diet are effective therapeutic options in the early stages of insulin resistance.
High-fat diets are atrocious for insulin sensitivity.