In the end a lot of this is taste but:

Indent based flow control is a massive fail right from the start.

By some metrics, Python is the most popular programming language in the world and has been for a long time. So the thing that's a massive fail right off the bat to you, is actually the lowest a barrier of entry gets. Simple GDScript starts out as basically a Python variant.

I've been a professional programmer for 20 years (not in Python) but I've never used C# and all the bits of it I've seen seem awful, like Java which I have used and hated every time. If Godot had been a C# system it would have been harder for me to get into it and enjoy it. Personally. I think the only reason they offer it at all is to try and appeal to Unity migrants.

GDScript was meant to be a gentle introduction. You've got a thing on the screen, and you want to just quickly move it when the cursor moves and it takes a few lines of code, every one of which is just about that intent. The way it has built-in syntax for addressing the hierarchy of nodes specifically encourages their use and incentivizes their design philosophy. I love functional programming typically, but I don't think I've ever missed having a lambda in GDScript, because I'm making a videogame, not transforming a data pipeline.

Things like class and class_name are a smidge confusing, sure, but both are essentially advanced features. The normal way to make a class is to make a file, and the normal way to refer to it is to include it in your node hierarchy, or by filename in a weird case.

So the GDScript is meant to be "progressive", like a scripting language, where it's a few lines to add a bit of interaction to something, where every line is there for a reason rather than boilerplate, and there are more advanced features if you need them, but they're not front and centre most of the time. I think most of the problems you've mentioned are just baggage you've brought from your other languages, but aren't needed here.

So anyway, it's taste. But fundamentally, I think the thing we disagree on most is "we don't need more languages, we need better languages". I get that, but disagree. I love a domain specific language, I love a purpose-built feature, and would much rather see a garden of variety rather than have everything be Java and C#.