Some are valid, some are not!
1. Consistent window decorations - Legacy apps don't even have a consistent UI. So what advantage will I get having consistent WM decorations only? About apps will be able to do their own decoration which is inconsistent, apps like Chromium do that, and they don't require CSD to be able to do it.Also, you worry a lot about GTK apps in KDE. It's not possible to get a consistent experience in apps which are written in 2 entirely different toolkits. You might add that thing to kde-gtk-config, that would be sufficient.
In my opinion, KDE should focus providing a consistent experience for their own apps. You cannot always provide consistent experience with apps written in GTK, EFL etc. I don't say it's a bad thing, but it always puts more work which could be spend otherwise to make KDE better.
2.. Close hung applications - valid
3. Consistent user actions between decoration and task manager - again, you can easily achieve this with KWIN I guess. I won't expect all window managers to provide same set of features. I think KWIN is the KDE window manager, thinking too much about other window managers will give no profit to KDE.
4. User actions menu in general , 5. One place to configure mouse actions, 6. Different settings for the maximize button, 9. Adding additional buttons, 10. Remote X-Clients, 12. Tooltips - now I'm not even sure, if you are talking about KDE apps with CSD or only GTK apps.
7. Differentiating between window and decoration - I'm eager to know about the use cases.
8. Window Shading - valid
11. Shadows - Internal technical stuff!
13. Netbook mode - aren't the decorations already hidden?
14. Accessibility - seems valid
15. Window Tabbing - valid
16. Numbering in window title - valid (but I guess descriptive window titles are better than numbered titles)
17. Forcefully adding buttons - use cases? for example, if the app doesn't support maximizing, what gain will be there by adding a maximize button?
18. Changing themes - again, I don't think you have to care a lot for theming of GTK.
I don't say you are wrong. But as I see, you think a lot about how a GTK app will look in KDE with CSD. Most of the points you mention is about GTK apps in KDE. I would prefer to consider those points with KDE apps in KDE.
GTK is a GNOME thing. They made their decision, and they want that. The responsibility of KDE devs is to improve KDE (I must say you do a great job), but what's the point of caring so much for an alien toolkit? Why don't you care the same for EFL, and all other toolkits that exist? Won't it be better to concentrate on KDE's own apps? Perhaps if you had unlimited developers you could think about every one. But you don't. Resources are limited. So better spend them wisely.
Again, don't take it as an offence. KDE has long reputation of making it consistent with GTK apps, but sadly, how hard you try, you cannot convince GNOME devs. Cause they care about how GTK apps look in GNOME, not about how they look in several other desktop environment. I would say KDE should also do the same. Because resources are limited, and they are better spent this way.