[Media Router] Reland: MediaRouter interfaces with stub implementations.

Reland of "MediaRouter interfaces with stub implementations."

Original patch: crrev.com/1020743003.

The GN file is changed so that the tests are no longer depending on the target that was causing link errors on debug GN builds.

TEST=git cl try + mac_chromium_gn_dbg + win8_chromium_gn_dbg

[Media Router] MediaRouter interfaces with stub implementations.

- Updated Media Router data classes.
- Added MediaRouter interface with a stub MediaRouterImpl class.
- MRImpl is a KeyedService and is instantiated via
  its BrowserContextKeyedServiceFactory implementation.
- Added MediaRoutesObserver / MediaSinksObserver interfaces which are
  used by MediaRouter API.

BUG=461815, 464199
Committed: https://crrev.com/2ae6fd35a8ce0dbaebd478efa342474adea63378
Cr-Commit-Position: refs/heads/master@{#324662}

[imcheng, 2015-04-10 06:10:45]

imcheng@chromium.org changed reviewers:
+ imcheng@chromium.org, kmarshall@chromium.org

[mark a. foltz, 2015-04-10 16:48:03]

mfoltz@chromium.org changed reviewers:
+ mfoltz@chromium.org

[mark a. foltz, 2015-04-10 16:48:03]

lgtm

[mark a. foltz, 2015-04-10 16:49:20]

On 2015/04/10 16:48:03, mark a. foltz wrote:
> lgtm

FYI, I've usually seen a re-land attached to the original issue (by manually
setting the issue # with git cl), so the entire history of the patch is in one
place.

[xhwang, 2015-04-10 16:59:23]

On 2015/04/10 16:49:20, mark a. foltz wrote:
> On 2015/04/10 16:48:03, mark a. foltz wrote:
> > lgtm
> 
> FYI, I've usually seen a re-land attached to the original issue (by manually
> setting the issue # with git cl), so the entire history of the patch is in one
> place.

Usually in media/ we don't reland CL in the same/original issue to avoid
confusion. I searched in my inbox for "Reland" and it seems this is the common
practice.

Also for the title, usually we do 

Reland "Origianl CL Title"

or 

Reland: Origianl CL Title

instead of using the crrev link in the title.

Here are some examples:

https://codereview.chromium.org/845033002/
https://codereview.chromium.org/655603003/
https://codereview.chromium.org/140823012/

[xhwang, 2015-04-10 17:02:22]

xhwang@chromium.org changed reviewers:
+ xhwang@chromium.org

[xhwang, 2015-04-10 17:02:22]

Also, I assume your PS1 is the original CL (crrev.com/1020743003), right? It's a
good practice to upload the original CL first (and make this explicit in the PS
title), then upload updated CL in a separate PS. This way, it's clear what has
been changed since the original CL.

Here's an example:

https://codereview.chromium.org/140823012/

Note the "Final patchset from original CL" part for PS1.

[imcheng, 2015-04-10 17:02:42]

imcheng@chromium.org changed reviewers:
- xhwang@chromium.org

[imcheng, 2015-04-10 17:05:34]

Updated title.

Yes, PS1 is the final PS from the original patch. I should've labelled it as
such in the PS title.

[imcheng, 2015-04-10 17:06:18]

The CQ bit was checked by imcheng@chromium.org

[commit-bot: I haz the power, 2015-04-10 17:07:10]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1076973002/60001

[commit-bot: I haz the power, 2015-04-10 17:59:30]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2015-04-10 18:00:23]

Patchset 4 (id:??) landed as
https://crrev.com/2ae6fd35a8ce0dbaebd478efa342474adea63378
Cr-Commit-Position: refs/heads/master@{#324662}