I can think of cases where there is no website available as an alternative to an app and yet the app can work on both pc and mobile. There's also the case for xbox though I have no idea how you use apps there
Take an example of a simple photo filter app. Can it be done via a website? Maybe. Will users like to keep starting their browser, go to the site, upload photos to apply filters and then download them? Or would they prefer an app with inbuilt filters that they can open at anytime without the hassle of using a browser and using data?
Your example fully supports my point rather than the one you're trying to make.
Let's assume nobody will ever build or want to use a web based photo editing solution. That's off the table. Then we must consider that there are already a gazillion Win32 based software packages for photo editing. If someone is a bit more professional about photo editing, they are likely using Photoshop or something similar on their desktops/laptops. If they aren't they'll do it on their iOS or Android smartphones. I can't see a lot of people wanting to do photo editing on their Xbox (worst usability ever).
So, given this situation, what is it that would convince developers it's worth porting an existing software package to the UWP?
Nothing...
That costs a lot of money and if the developer is also targeting OSX and/or Linux, it's likely to make things a lot more complicated than they already are. If a developer is going to make an investment into the UWP, there needs to be a return on it. Explain to me how UWP provides that return? Only if you can do that do you have a point. Otherwise developers won't care.
The best argument you could make is that someone might be developing a brand new app, in which case you can flip the question on its head and ask "why not"? You're going to be investing a lot of money anyway so why not make it an UWP app? If the developer is planning to only ever target Windows I guess that's a ligitimate argument. That's just almost never the case. Most developers want to also target OSX. In such situations other UI technologies are employed to make it easier to target multiple systems (like Qt). At that point, at least from a consumer's point of view, the UWP just reverts back to being the same old solution to a problem too few people care about... bringing desktop software to W10M.
If you can make a compelling case by explaining what consumer related problem the UWP solves that developers actually care about, I'll gladly change my mind. I just don't see how you could.