The decl checker was effectively not being run on these because we weren’t typechecking the PBD and typechecking the VarDecl itself is basically a no-op.

Macro expansions are now treated like a part of the source file they belong to, for purposes of the “second declaration is the one that’s diagnosed” rule. This helps stabilize a behavior that was easy to perturb.

CustomAttr backs four different features, each of which requires a different behavior in `@abi`:

• Global actors: Permitted (and permitted to vary) since they can affect mangling
• Result builders: Forbidden inside an `@abi` since they have no ABI impact
• Property wrappers: Forbidden both inside an `@abi` and on a decl with an `@abi` since it’s not clear how we would apply `@abi` to the auxiliary decls
• Attached macros: Forbidden inside an `@abi` since an ABI-only decl has no body, accessors, members, peers, extensions, or (currently) conformances

Implement these behaviors (outside of `ABIDeclChecker` since they can’t be described there).

Macro-related tests are not included in this commit; they require matching swift-syntax changes which are being negotiated.

SwiftSyntaxParser is already doing this, and we already diagnosed it in Sema anyway, so we’re just moving that diagnostic earlier so the ASTGen testing mode is happy. Also adding compiler tests for it.

Macro-related tests are not included in this commit; they require matching swift-syntax changes which are being negotiated.

It’s not clear how `@abi` would apply to the auxiliary decl used for `lazy`.

Inlinability doesn’t affect the mangling except in function specializations, which are applied after the fact and should never mangle in information from an ABI-only decl. That means we can simply ban these from `@abi` instead of inferring them.

Also adds some assertions to help double-check that SIL never tries to directly mangle or retrieve inlinability info from an ABI-only decl.

It has indirect effects on the accessors, so it shouldn’t matter, but we can defensively redirect the query to the API counterpart anyway.

This was the last `InferredInABIAttr` attribute, so we can now remove all of the infrastructure involved in supporting attribute inference.

When a language feature is used inside an `@abi` attribute, we should behave as though it was used on its counterpart. This was already half-implemented—we ensured the counterpart would use the feature—but we didn’t make the ABI decl aware that the counterpart was its parent for feature detection purposes. As a result, we would print `#if` inside the `@abi` attribute, which isn’t valid.

Additional tests for the previous commit “Diagnose CustomAttrs as needed in `@abi`”.

Additional tests for the previous commit “[Legacy parser] No freestanding macros in `@abi`”.

Per a comment in the review thread that the comma in a multi-variable pattern binding makes it look like the `@abi` attribute has several arguments.

This includes changing the feature name so that compilers with the experimental feature don’t accidentally pick up content that only works in the final version.

Resolves rdar://150065196.