Which is why I included a more broad definition in my discussion. If I like somebody until I find out that they have some characteristic that is a dealbreaker for me in a relationship, that doesn't inherently make me a bigot. Because this is a much more complicated issue. My point is that "liking somebody" cannot be used as a standard for bigotry. That's not a reasonable standard.
Let's say there's a woman I'm really attracted to, I've talked to her online, and then I meet her in person, and her voice grates on me. It is the most annoying thing I've heard. I know that it would not be fair of me to try to suppress this to date her, because there's another person who will not mind out there. And the same holds true of the transpeople in this scenario.
Stating something doesn't make it so. And in this case you're stating that other opinions which are not exactly minority opinions are inherently bigoted and awful. In a practical sense, that isn't a position that's going to gain you much traction except for with the extreme edges of society.
Also you're the one who declared that it was transphobia, and further attempted to argue that people who have a differing position are by definition transphobes.
Being trans is the part of your character that makes you a woman in this case. And that fact is significant here because that is a thing that would bother many romantic partners as has been shown by survey here. Being unrealistic about that isn't going to help or improve matters.
In 99.95% of all cases... It is not. The frequency of actual intersex births is something like 1 in 1500 to 1 in 2000. So no, that is not a good standard to use to simply throw out the 99.95% of all cases where that is indeed a valid measure. That passes every test for statistical certainty you could throw at it. Even if you're including the trans population estimates, that only brings you to around 99.45% of all cases. That's still not enough to throw out biological gender as a model.
Edit:
Using biology (chromosomes) and not being trans would work in roughly 98% of all cases. Therefore that's probably the most accurate method you're going to get.
It's like 1.2 percent of the population in the most generous estimates, that's far less common than natural redheads. At least where I live. If you include Asian and African populations you may drop the number of redheads, but that seems to be not really that relevant here.