Handle async functions in traits

[braw-lee]

Fixes #2785
I opened this pull request because I have some doubts.
So I created a switch-case for async keyword when reading trait items. I generate an error (I know its not the right way) and return nullptr.
The returning nullptr generates additional errors that results in the test case failing.
The correct way to generate error would be to add check in ASTValidation::visit (AST::TraitFunctionDecl &decl) function but it won't work right now because we never made the AST node.
To solve this problem I think we need to implement parse_trait_async() and to implement this function we need to make an AST node AST::TraitItemAsync.
But this node will be pretty much useless as async isn't allowed inside traits so I don't know if this is the right way to approach this problem
@CohenArthur Is there any other way to solve this problem, I am sorry if I failed to reach an obvious solution, lemme know.

[P-E-P]

function but it won't work right now because we never made the AST node. To solve this problem I think we need to implement parse_trait_async() and to implement this function we need to make an AST node AST::TraitItemAsync.

We usually only create a node for a given syntax tree component, the async keyword gives us information but cannot have any "child", it contains the asyncness information. Function declarations store the asyncness in function qualifiers.

Even though async should probably be rejected in the ast validation stage, the compiler struggles at an earlier stage (during parsing) per your example. It expects some token and find another, it already covered all possible cases and expects a macro call (!).

This means we should modify the parsing stage to accept the async token, store the asyncness information in the function qualifiers and reject an async function declaration in traits at a later stage during ast validation probably around here like it has been done with const functions.

Hopefully this is clearer ?

[braw-lee]

yes, this is clearer. Thanks for help.

[braw-lee]

@P-E-P is this PR good to merge?

[CohenArthur]

looks great to me! thanks!

[CohenArthur]

thanks a lot!