A couple points:
One of your broader points seems to be that providers with negative countertransference towards gay and trans people can cause harm. This is obviously true, but I would also contend that it is a good deal less relevant in the context of providing gender-affirming treatments. The reality is that physicians tend to gravitate to treating populations that they fundamentally like. My experience has been that the people who are doing evaluations for gender dysphoria tend not to be people with simplistic or negative views about gender. They tend to be people who are pretty open about the nature of gender. That is why I brought up the fact that positive countertransference can be a problem—because those are the people who are primarily treating trans people.
I will stay away from the discussion of non-binary gender identity for now, because that is a very complicated issue. All I will say is that I think there are people with legitimate dysphoria due to non-binary gender identity but, in practice, assessing it is very complicated. My experience has been that most of the people identifying as non-binary do not present with a classic gender dysphoria picture and do not have the history that usually accompanies gender dysphoria. Not every trans person with gender dysphoria has a history of cross-gender play behaviors and intensifying body dysphoria as they approach and go through puberty, either, but this seems significantly less common with people identifying as non-binary. My general experience has been that these individuals often develop a sense of their identity later, have less clear ideas about exactly what phenotypic changes would resolve their distress, and have higher rates of comorbid mental illness. In many cases, there's not really a clear answer regarding how to best help these patients. These decisions are made even more difficult by the fact that the idea of non-binary gender identity is relatively recent. Yes, you can say that non-binary people have existed historically, but clinical and research attention to the issue is relatively recent. There really is a paucity of good information about this population and their prognosis with various interventions. By comparison, we have a lot more solid data for binary trans patients.
You make the point that someone can have mental illness and gender dysphoria. I don't disagree with this. I explicitly mentioned this as a possibility in my initial post. If you're making some sort of political point by raising concern that those seeking inappropriate interventions shouldn't be used to deny care (on a systemic or governmental level) to people with legitimate dysphoria, I don't really disagree with this either. I will say, though, that as an individual provider it is your responsibility to figure out who would likely benefit from medical interventions and who would not. Inherent to evaluating someone for any condition that might indicate a certain intervention is some sort of gate-keeping function. Being a good doctor inherently means being willing to deny people inappropriate treatments. That is an essential part of medicine. The fact that most patients seek treatments for appropriate reasons shouldn't really factor into the way you approach patients. The reason is that this is a population-level statement about motives, rather than an individual statement about the patient you are actually seeing. As a general rule, most people asking for antipsychotics are probably asking for it for appropriate reasons. That doesn't mean that you shouldn't be trying to figure out if they're just trying to get high on Seroquel or sell it on the streets. I also think that it is generally good clinical practice to always look for the thing that you don't expect. If someone seems totally believable to you, you should force yourself to at least consider the possibility that they're malingering. If someone seems classically borderline, you should at least consider the possibility that what you're observing is actually an affective illness, etc. Same thing here. As a physician, you're supposed to be looking out for the minority cases rather than just doing a cursory assessment and playing the odds.