std: Stabilize TypeId and tweak BoxAny

[alexcrichton]

This commit aims to stabilize the TypeId abstraction by moving it out of the
intrinsics module into the any module of the standard library. Specifically,

• TypeId is now defined at std::any::TypeId
• TypeId::hash has been removed in favor of an implementation of Hash.

This commit also performs a final pass over the any module, confirming the
following:

• Any::get_type_id remains unstable as usage of the Any trait will likely
  never require this, and the Any trait does not need to be implemented for
  any other types. As a result, this implementation detail can remain unstable
  until associated statics are implemented.
• Any::downcast_ref is now stable
• Any::downcast_mut is now stable
• BoxAny remains unstable. While a direct impl on Box<Any> is allowed today
  it does not allow downcasting of trait objects like Box<Any + Send> (those
  returned from Thread::join). This is covered by Replace current quasi-subtyping relationship for object types with coercions #18737.
• BoxAny::downcast is now stable.

[alexcrichton]

r? @aturon

[rust-highfive]

r? @aturon

(rust_highfive has picked a reviewer for you, use r? to override)

[seanmonstar]

Typo limitation

[aturon]

OK, I'm happy with this, but want to get @nikomatsakis's opinion as well.

[nikomatsakis]

So obviously the key question is this function, right? I'm roughly willing to go with this, but I'd like to get a better understanding of whether we could get by with some kind of bound here.

[nikomatsakis]

On Thu, Jan 15, 2015 at 10:51:22AM -0800, Aaron Turon wrote:

OK, I'm happy with this, but want to get @nikomatsakis's opinion as well.

After discussing with @reem in IRC, it seems like we want to move to a
design where there is a Reflect trait that is automatically
implemented for all types:

trait Reflect : 'static { }
impl Trait for .. { }

TypeId would thus require a Reflect bound

@flaper87 is actually pretty close to having the .. stuff implemented in full, I think,
so this could plausibly land very soon.

In the meantime it might make sense to keep of() as unstable?

[alexcrichton]

r? @aturon

I've marked TypeId::of as #[unstable] for now until the marker trait is implemented

[aturon]

Sidenote: the (currently stable) blanket impl for Any will likely also grow a reflect bound. This is unlikely to be a breaking change in practice.

[reem]

It will actually require a Reflect bound, since it will use TypeId::of.

[alexcrichton]

@bors: r=aturon 80d73e9

[alexcrichton]

@bors: retry

[alexcrichton]

@bors r=aturon e1e0450

[alexcrichton]

@bors: r=aturon 70f165

[bors]

🙀 You have the wrong number! Please try again with 70f7.

[alexcrichton]

@bors: r=aturon 70f7165

[bors]

⌛ Testing commit 70f7165 with merge 65b61ff...

[bors]

☀️ Test successful - auto-linux-32-nopt-t, auto-linux-32-opt, auto-linux-64-nopt-t, auto-linux-64-opt, auto-linux-64-x-android-t, auto-mac-32-opt, auto-mac-64-nopt-t, auto-mac-64-opt, auto-win-32-nopt-t, auto-win-32-opt, auto-win-64-nopt-t, auto-win-64-opt