Reland "Add --serial-id option to host_forwarder."

Reland "Add --serial-id option to host_forwarder."

This relands r209945.

r242846 slightly changed the CLI <-> device daemon message sequence which made
the following check fail in case the device daemon was started but already
running on the device: device_forwarder_main.cc:10: CHECK_GT(bytes_read, 0).

This was actually caused by an unexposed long standing issue in daemon.cc.
ConnectToUnixDomainSocket() is performing a hand check to make sure that the
CLI is connecting to the forwarder daemon. While doing this hand check, rather
than only reading the expected fixed-size message, the function was reading all
the available bytes in the socket thus consuming the client data.

BUG=242846
R=bulach@chromium.org

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=210273

[Philippe, 2013-07-03 13:41:57]

FYI, you can diff against patch set 1 to see the fix.

[bulach, 2013-07-03 13:55:00]

lgtm, good catch! nit: please change the title to "Reland" :)

[Philippe, 2013-07-03 14:01:01]

On 2013/07/03 13:55:00, bulach wrote:
> lgtm, good catch! nit: please change the title to "Reland" :)

Thanks Marcus!

[commit-bot: I haz the power, 2013-07-04 15:51:52]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/pliard@chromium.org/18635005/21001

[commit-bot: I haz the power, 2013-07-04 16:21:42]

Sorry for I got bad news for ya.
Compile failed with a clobber build on win_x64_rel.
http://build.chromium.org/p/tryserver.chromium/buildstatus?builder=win_x64_re...
Your code is likely broken or HEAD is junk. Please ensure your
code is not broken then alert the build sheriffs.
Look at the try server FAQ for more details.

[Philippe, 2013-07-05 07:42:07]

Committed patchset #4 manually as r210273.