Speaking to the second item mentioned in the original post. How is someone supposed to know if a particular tree was planted or sprouted on its own accord? You don’t. It’s not a dicotomy where the choices are (1) this tree was planted by humans and (2) this tree was not planted by humans. The true dicotomy is (1) I know that this tree was planted by humans and (2) this tree may or may not have sprouted on its own accord. There can be evidence or knowledge of a tree having been planted, but there can be no evidence or knowledge of the opposite. Any one tree in a forest could be there because someone pulled a seed out of their pocket and stuck it in the ground. iNat defaults to “wild” if the box is not checked, but this conveys a false sense that it is known with any confidence that it is indeed wild. In practical reality, the true dicotomy must be as I described above.
Speaking to the third issue: it being unfair or uneven comparing plant designation of cultivated with animal designations of captive because animals can move and plants cannot. Why do we need this to be a fair or even comparison? It just isn’t and it shouldn’t be expected because the types of evidence are different. No dataset or study will be negatively impacted because of this difference in our ability to make these judgments in animals vs. plants. A dog on a leash or a cat sitting in someone’s lap in their apartment are clearly captive–and thus that box can be checked. A dog or cat in the woods cannot be determined to be wild (out there surviving on it’s own) or recently escaped from captivity–and thus that box cannot be checked.