Added support for VectorIntoWasmAbi and VectorFromWasmAbi

[eneoli]

In this pull request (see also: this related issue), the capability to return a Vec<SomeType> was introduced to wasm-bindgen, where SomeType has the #[wasm_bindgen] attribute.

This PR further extends this capability to support returning Vec<SomeType> where SomeType derives Tsify. This is done by implementing VectorIntoWasmAbi and VectorFromWasmAbi introduced in the linked PR.

Two new attributes were added to control whether the above traits should be implemented.

Example:

#[derive(Serialize, Deserialize, Tsify)]
#[tsify(into_wasm_abi, from_wasm_abi)]
#[serde(tag = "kind", content = "value")]
pub enum Pet {
    Cat(String),
    Dog(String),
}

#[wasm_bindgen]
pub fn return_pets() -> Vec<Pet> {
    vec![Pet::Cat("Pumpkin"), Pet::Dog("Cookie")]
}

#[wasm_bindgen]
pub fn read_pets(pets: Vec<Pet>) {
    // ...
}

[eneoli]

@siefkenj Sorry for the noise. I had an old clippy installed that did not detect the dead code. This also probably explains why the CI did not nag when this code was added first. (The dead code is unrelated to the PR.)

[siefkenj]

Thank you for the contribution! I will review it as soon as I can :-)

[siefkenj]

Is there a reason to put these behind a flag? Would it cause problems if into_wasm_abi implied vector_into_wasm_abi? (I would think that would be the default expected behavior)

Could you please add some tests to this PR?

[siefkenj]

Were these comments not being used? Why was this field removed?

[eneoli]

Correct, the static code analysis (clippy) reported that this field was never read. That was also one of the reasons the CI failed.

[siefkenj]

Thanks for your effort!

[eneoli]

Thank you for merging my pull request!

Could you please also publish a new release that includes these changes?

[siefkenj]

Done :-)

[Pantamis]

Thank you so much !

This is really great stuff ! I was just missing this XD