Roughly speaking, there are approximately 18 billion super awesome things that Sublime could be doing out of the box without requiring a plugin (give or take a few billion either way).
One of the things that keeps Sublime small, tight and fast is that the developers carefully pick and choose what they “officially” support by building it directly into the core, and for everything else they provide an extension API so that everyone can make Sublime do what they want (within some limitations) without extra baggage for people that don’t need/want that kind of functionality.
The flip side of this is applications like IntelliJ, Visual Studio etc that provide all sorts of functionality out of the box. That creates a core product that’s more complicated to maintain for the developers because there is more code, makes the product bigger and also (to some extent) makes it slower.
If you happen to be in the camp that wants all of that functionality, you’re happy because it’s there, and if you don’t, you’re sad because it’s there.
The Sublime devs seem to be very carefully considering what they think should appear in the core versus what can be added via extension packages and plugins so that they can keep things small and fast in the core. Much like my wife, their reasoning is sometimes clouded in mystery and/or may not make as much sense as one might like, but it is what it is.
Essentially it’s a really slippery slope; everyone has their “Sublime is cool but it just needs this one extra feature and it would be the best”; there are a lot of users of Sublime, so if everyone got their one item in, we’d all be unhappy.
All of this said, we don’t really have any concrete list of what they will or will not implement, only what they have and have not implemented yet.