Show function <name>() { [native code] } for built-in classes

Show function <name>() { [native code] } for built-in classes

The existing logic would show the full source for all classes, even
built-in ones.

R=arv@chromium.org,dslomov@chromium.org
BUG=

Committed: https://crrev.com/a34bbef3146902e5268ce899444cf011b501b300
Cr-Commit-Position: refs/heads/master@{#28151}

[arv (Not doing code reviews), 2015-04-29 15:05:02]

Can this be tested?

At first I wasn't sure using function would be better than using class but upon
further thought function is better because it does not give away implementation
strategy.

[domenic, 2015-04-29 15:15:42]

On 2015/04/29 at 15:05:02, arv wrote:
> Can this be tested?

Not easily ... right now none of the built-ins use `class`. The only strategy
that comes to mind in that case is to run a separate v8 build with extra
built-ins (possible these days through the v8_extra_library_files gyp variable)
and run tests against that. Better ideas welcome.

I was less bothered by this after realizing there are no tests for the existing
behavior for non-classes. (Except one indirect one for poisoned caller, IIRC.)

> At first I wasn't sure using function would be better than using class but
upon further thought function is better because it does not give away
implementation strategy.

Right, this plus consistency with all of the built-ins and web platform classes,
was my thinking.

[Dmitry Lomov (no reviews), 2015-04-30 08:47:24]

lgtm

[domenic, 2015-04-30 09:06:14]

The CQ bit was checked by domenic@chromium.org

[commit-bot: I haz the power, 2015-04-30 09:06:40]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1112113002/1

[commit-bot: I haz the power, 2015-04-30 09:29:38]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-04-30 09:29:50]

Patchset 1 (id:??) landed as
https://crrev.com/a34bbef3146902e5268ce899444cf011b501b300
Cr-Commit-Position: refs/heads/master@{#28151}