Update harmony ObjectProtoToString() to 2/2/2015 spec

Update harmony ObjectProtoToString() to 2/2/2015 spec

Applies to API implementation of O.p.toString

- Removes special handling of non-string @@toStringTag values (use builtinTag)
- Removes special handling of @@toStringTags which match [[Class]] names (remove ~ prefix)
- Add tests for non-string @@toStringTag values in test-api.cc

BUG=v8:3502
R=arv@chromium.org
LOG=N

Committed: https://crrev.com/b5c43674f998e5660f110271fe19fc953aee9dcb
Cr-Commit-Position: refs/heads/master@{#26415}

[caitp (gmail), 2015-02-03 15:39:18]

PTAL, api portion (I'm not sure if the description communicates this very well)

[caitp (gmail), 2015-02-03 15:39:19]

PTAL, api portion (I'm not sure if the description communicates this very well)

[caitp (gmail), 2015-02-03 15:39:50]

On 2015/02/03 15:39:19, caitp wrote:
> PTAL, api portion (I'm not sure if the description communicates this very
well)

oh rietveld did you just send the same comment twice >_-

[arv (Not doing code reviews), 2015-02-03 16:06:38]

LGTM

Didn't the cc test have a case for ???

[arv (Not doing code reviews), 2015-02-03 16:07:14]

On 2015/02/03 15:39:19, caitp wrote:
> PTAL, api portion (I'm not sure if the description communicates this very
well)

The description could mention that this is the cc part.

[caitp (gmail), 2015-02-03 16:08:05]

On 2015/02/03 16:06:38, arv wrote:
> LGTM
> 
> Didn't the cc test have a case for ???

apparently not, but it would have just been deleted I guess. I can add one to
assert that it just uses the class name instead

[caitp (gmail), 2015-02-03 19:16:54]

On 2015/02/03 16:08:05, caitp wrote:
> On 2015/02/03 16:06:38, arv wrote:
> > LGTM
> > 
> > Didn't the cc test have a case for ???
> 
> apparently not, but it would have just been deleted I guess. I can add one to
> assert that it just uses the class name instead

Added some extra tests, does this still look ok?

[arv (Not doing code reviews), 2015-02-03 20:38:53]

Still LGTM

[caitp (gmail), 2015-02-03 20:40:53]

On 2015/02/03 20:38:53, arv wrote:
> Still LGTM

Thanks, cq'in

[caitp (gmail), 2015-02-03 20:40:58]

The CQ bit was checked by caitpotter88@gmail.com

[commit-bot: I haz the power, 2015-02-03 20:41:08]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/892393003/40001

[commit-bot: I haz the power, 2015-02-03 21:05:14]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2015-02-03 21:05:39]

Patchset 3 (id:??) landed as
https://crrev.com/b5c43674f998e5660f110271fe19fc953aee9dcb
Cr-Commit-Position: refs/heads/master@{#26415}

[caitp (gmail), 2015-02-03 21:29:04]

On 2015/02/03 21:05:39, I haz the power (commit-bot) wrote:
> Patchset 3 (id:??) landed as
> https://crrev.com/b5c43674f998e5660f110271fe19fc953aee9dcb
> Cr-Commit-Position: refs/heads/master@{#26415}

there's a problem in debug builds... I'll work on a fix but does this need to be
reverted?