The study you quote was funded by Virta Health Corp. a health startup that advertises "The Virta treatment [that] reverses type 2 diabetes". They use this study as their main advertising item, quoting it extensively on their site, calling it "groundbreaking" etc etc.
Besides the fishy smell emanating from this, and to address your comment directly, the quoted study compared the proposed treatment against "usual care", i.e. ordinary diabetes treatment.
More specifically, the study did not directly compare the proposed treatment to other dietary or lifestyle interventions and particularly not to calorie restriction diets. It does, however, quote the results of such studies:
The CCI [Continuous Care Intervention] reduced HbA1c by 14 mmol mol−1 (1.3%) at 1 year. HbA1c reductions up to 7 mmol mol−1 (0.6%) via intensive lifestyle intervention [25] and 11 mmol mol−1 (1.0%) via an energy-restricted low-carbohydrate diet with partial food provision delivered via an outpatient setting [26] were previously reported.
So the difference between this and one selected calorie restriction diet (which was also a low-carbohydrate diet) was small (3%). These are just two studies, so it's hard to draw a conclusion.
Anyway, I'm not a doctor or a biologist so I don't know just by looking at the studies to say what is significant and what isn't. For instance, sulfonylurea treatment was reduced in the "CCI" group, like you note, but that group already had ~24% lower use of sulfonylureas than the baseline to begin with. What does that mean? And what does it mean that people took less medicine at the end of a study? Did they absolutely need all the medicine they were taking at the start? Do you know how to answer those questions? Because I certainly don't.
A single study, quoted by someone who isn't a doctor to people who also aren't doctors over the internet, doesn't really mean anything.