That's too easy as an explanation. "We can do whatever we feel like as each of our customers will have different wishes anyway". No.
There's some pretty well established things in the world of "flagship DAWs" and it's possibly a decent idea to have a good amount of them covered. And as these have been mentioned in this thread: Region gain handles and proper time stretching for example. They're pretty much kind of an "industry standard" by now. Same goes for a half decent sample editor, especially when you stopped developing what you have 20 years ago already. Etc.
Similar things might be true for, say, ARA integration. Yes, I perfectly understand why it's not working in silicon-native mode anymore, but in case it's not, 3,5 years should be a good enough amount of time to adress the issue. Ideally, they would've thought this through before changing the way plugins are called up. Apparently, that just didn't happen.
All of this isn't exactly about new features (for instance, region gain handles aren't exactly a new feature, just another way to present what's already there), same with my ongoing complaints about them to completely break an entire section of zooming functionality.
Hence, while nobody exactly minds whatever new things they're throwing at us, especially among longstanding Logic users, there seem to be quite some people wishing for a really intense housekeeping session. Personally, I'd happily pay for an upgrade just adressing all these issues.