Update FFI documentation for new enum layout with smaller/selectable discrminants.

[jld]

@alexcrichton points out that, when PR #9006 (currently being inspected by bors) lands, some of the FFI documentation may become wrong if it talks about enum representations. (Or, if it's not wrong, it should probably be extended.)

[huonw]

Triage; still a bug. I guess it should/could be added to http://static.rust-lang.org/doc/master/tutorial-ffi.html#interoperability-with-foreign-code .

[steveklabnik]

"However, there are currently no guarantees about the layout of an enum."

This should be closed.

[alexcrichton]

While we don't normally provide guarantees, the #[repr] attribute (which I believe this is talking about) is used for interoperability with foreign code, which is one location where we need to be strict about the representation of enums.

[steveklabnik]

Rust guarantees that the layout of a struct is compatible with the platform's
representation in C only if the #[repr(C)] attribute is applied to it.
#[repr(C, packed)] can be used to lay out struct members without padding.
#[repr(C)] can also be applied to an enum.

This seems to be fine.