I was diagnosed as Type 2 at age 40, and two months later (at age 41) re-diagnosed as Type 1 based on antibody tests. I didn't eat or "sit" my way into either diagnosis. I've always eaten well, and I've always been active, fit, and healthy as both a child and an adult.
I think the whole child vs. adult split up until the 1980s was simply enough a misunderstanding based on lack of knowledge. We understand the genetics of Type 1, Type 2, MODY, MIDD, and other forms of diabetes mellitus now in a way that we didn't (and couldn't) know before that period. We now know that Type 1 isn't as simple as an autoimmune attack on beta cells during childhood. Type 2 isn't as simple as the "natural" result of obesity and overeating in adults. It's all far, far more complicated than that oversimplification.
I think the increase in adults is a change in terminology and diagnostic criteria. In the good old days, you were either "insulin-dependent" or "non-insulin-dependent." There was also "insulin sensitive" and "insulin resistant." There have been "fat" and "skinny" diabetics. I think what has changed in the last thirty years is pretty simple: better genetic testing and understanding of genomic science (see Trialnet); more and better antibody-tests that can be used to diagnose "Type 1" DM; a greater focus on metabolic disorders in general as the population ages and struggles with obesity-related disorders.
The last one is somewhat controversial, since obesity and Type 2 are correlated, but not necessarily in a causal relationship with each other (or at least not one we understand). The side-effect of a greater focus on Type 2 and obesity is that more Type 2s are being diagnosed than ever before. Like me, if someone comes into a doctor's office and is over age 40 and/or obese, they automatically get diagnosed as Type 2 regardless of presentation. "It's statistically most likely." Well, I wasn't obese, I was over 40, but it turns out I don't have Type 2. I've heard as many as 20% of Type 2 diagnoses later turn out to be Type 1, MODY, or MIDD. But if an adult walks into a GP's office and has symptoms of diabetes, and their glucose metabolism is compromised, it's almost certain they'll be diagnosed as Type 2.
My old doctor (I found a new one) was young, intelligent, and didn't believe that adults could really be Type 1. He believed, and told me, that my "Type 2" was probably because I'd put on 5 lbs since I started graduate school in my mid 30s. I believed him, and felt full of self-loathing for having "done this to myself." I now know that even if I was Type 2, it wouldn't have been "my fault." But I thought I knew enough to know that people "at their way into Type 2" in adulthood. Or they were lazy... it is, after all, the "fat and lazy" disease.
Type 1, as far as I know, doesn't have anything at all to do with diet or exercise (or lack thereof), although both are important considerations when treating Type 1. Simply, genetic susceptibility, infection by various viral diseases, and then a follow-on autoimmune attack on beta (and more) cells in the pancreas are the "cause" of Type 1. Doesn't matter if the country is becoming more obese, sitting around more, or anything else. As a scientist, the only thing I can think of that would lead to a higher percentage of the population being diagnosed as Type 1 than in a previous generation are the following:
New diagnostic criteria (see above);
A higher incidence of the Type 1 genotype in the population (which may be happening as Type 1 is now treatable effectively in a way it wasn't 100 years ago--there is a likelihood that many Type 1s are passing their genes on now, when that is far less likely in the pre-exogenous-insulin past);
And a greater incidence of the types of viral infections that can precipitate an autoimmune storm leading to development of antibodies attacking beta cells (could happen as population density increases, but that's really outside of my level of understanding).