[META] Planned changes to AST structure

[Keno]

Discussing with @mlechu, I think there's a few near term changes we'd like to make to the core JuliaSyntax AST data structures. I'm opening up this issue so we have a place to track them all and discuss the overall plan. Particulars of any change can go to the appropriate issue or PR. I think the plan would be that we put them all in and then tag JuliaSyntax 2.0 for 1.13 to go with putting JuliaLowering in Base.

Planned changes (will amend list as more things come up).

1. Separate out "trivia" from "notation" #564
2. Remove the DOTOP_FLAG - it's hardly used at this point anyway (Remove DOTOP_FLAG from tokenizer and parser #568)
3. Find a different representation for the numeric bits - they are only used by two heads, but take up 8 bits in every syntax node. This seems a bit wasteful.
4. Remove individual syntax heads for unicode operators - replace with one head per precedence level (plus possibly a few special heads used elsewhere)

[Keno]

1. Find a different representation for the numeric bits - they are only used by two heads, but take up 8 bits in every syntax node. This seems a bit wasteful.

I did this change locally, but on further thought, I think we can and should just overlap these with the other head-specific flags in the top 8 bits of the flags. I don't see any other compelling reason to move these out.

[mlechu]

Overlap is fine, but I still think it would be bothersome to get or set the numeric flags in macros accepting or producing those forms

[Keno]

Is the same not true for the other head-specific flags? Do we have concrete examples of what macros will look like in the new system? I think that could help guide API design.

[mlechu]

Fair, it makes sense to wait until that's sorted out. Some examples of the current state of new macros are here