Unfortunately, I can’t tell you who is right and who is wrong here. However, when it comes to this sort of thing where people point fingers at each other, it’s important to display to them the opposite side so they can come together for some sort of resolution.
That said, as was said in one of the comments on the Qt bug:
we just have some widgets that could be used in a file manager, like QListView and QTreeView
Those are indeed the widgets that PCManFM-Qt uses. So it certainly seems that if anywhere, the problem exists in Qt, but I don’t know.
The problem with reporting something on Qt is exactly as you found: you need to show a problem with the way their code works, not how an application that uses it works, although it may be applicable. The best way to do this (although this is probably asking for more than you’re willing to do) would be to create a very simple program using those widgets and showing how its behavior is problematic.
Even if you can show that, all you would be showing is that it responds in a way that is different than alternatives. Different doesn’t mean that it’s wrong or bad. In fact, it may be entirely intentional that it function that way.
One thing to consider is that PCManFM-Qt should not be considered a 1:1 alternative to PCManFM anymore than LXQt should be considered a 1:1 alternative to LXDE. Though PCMan was instrumental in the initial porting, there was a wholly separate and unique team of people that put LXQt together relative to LXDE. They are completely different things and they use completely different toolkits that themselves have their own behaviors and requirements. The fact that LXQt does not work exactly like LXDE is not necessarily a bug. In most cases, it’s a feature.
Respectfully, I find this to be of such minor value that were I you, I’d just use Space and stop wasting my time. Think of how many times you have used it just trying to write about this so-called problem! 
