(I’ll start from a bit of nitpicking tbh, the more significant stuff is in the last part).
The “In-herit Alpha” toggle by itself doesn’t do anything
It does, but if the layer below is fully opaque, then inheriting alpha from below will just mean that all pixels are the “allowed” ones. If you do however paint an object and then shade it over a transparent background, you’d see that it does have an effect, exactly the same as in a group, because in Krita groups are used for composition (try using inherit alpha in a group, and then remove all other layers (including background) and ungroup the ones from the group with inherit alpha, then you’ll see it works just the same).
But that’s of course only a technical detail, most people don’t really need that.
This not only takes several actions
What if it was just easier to make? Like alt-clicking, for example?
it visually rivals actual groups that server a logistical purpose
In Krita groups are for both organization and composition (in PS, they are used only for organization, I believe - you can use Pass-through option in Krita to turn it on). Inherit alpha uses the fact that they affect composition. (And you can use other layers in the group besides the layers with inherit alpha turned on).
Ok most of that was nitpicking, but now the biggest thing which is the last one.
I have plenty of (genuine, they will help flesh out the design) questions. Have you used other types of masks in Krita? Do you know that in Krita you can use multiple masks on one layer? (I don’t think it’s possible in PS). How should it look like if a layer had five transparency masks? How should different kinds of masks be differentiated? (Is the different line between the layer thumbnail and the mask thumbnail the thing that will help the user understand which mask it is? If so, how is that better than a small icon if it’s such a small thing too?). Or maybe only transparency mask should be folded like that? And if so, why? Only the Colorize Mask is a bit of “odd one out”, all others are greyscale based just like transparency mask and have similar functions.
My opinion is that it could be sometimes useful to be able to fold like that, with an icon or alriiiight, a thumbnail to be able to select the mask quickly if you need to, but it would be only possible if there was just one mask. And it should be still possible to fold and unfold like it is now; the point of whole idea would be to be able to quickly access the folded mask (since now you have to unfold it first).
Frankly I don’t really see thumbnails as helpful there, if I see the layer, I know what the mask is, I usually paint on a mask without isolating it, so I don’t even know how it looks. And the screenshot you show as a comparison actually makes it worse visually for me, it looks much more messy.
In Krita it looks like another kind of group which is not ideal.
This is a parent-child (hierarchical) relationship within a tree, and the whole Krita’s layer stack is an ordered tree. Making it visually consistent is a pro in my opinion. But I dunno, maybe that’s just because I’m a programmer and I just know what a tree is (in technical context) and I’m comfortable with such structure. On the other hand, I would assume that people do understand the hierarchical structure Krita uses there (which is also something you can do with just groups in PS), at least no one ever complained about the structure itself.
Both putting masks next to the layer thumbnail and implementing clipping masks would break the ordered tree structure. The second one visually, the first one both visually (since it looks like a child relationship to the mask layer, but it’s on top instead of on the bottom) and potentially has bigger consequences in Krita’s compositing code (right now, every new node builds atop of everything below; clipping mask would have to be aware what exact node is just below it - if it was a child of the node, it would make sense to know about its parent, but it can’t be a child because it’s on top…).