Split DoHandshake into DoHandshake and DoHandshakeComplete.

Split DoHandshake into DoHandshake and DoHandshakeComplete.

The rv == 1 case was getting a little unwieldy. This should hopefully be
simpler. DoHandshake converts the error to a net error code and decides which
state to transition to. DoHandshakeComplete handles the result of the
handshake.

BUG=none

Committed: https://crrev.com/c4212c0b5cc733092b688b5eb1fa53eea1b15c08
Cr-Commit-Position: refs/heads/master@{#329518}

[davidben, 2015-05-08 22:29:16]

davidben@chromium.org changed reviewers:
+ rsleevi@chromium.org

[davidben, 2015-05-08 22:29:17]

This is actually a positive diffstat, but I think this is somewhat clearer and
more like how //net state machines normally work. What do you think?

https://codereview.chromium.org/1126113004/diff/1/net/socket/ssl_client_socke...
File net/socket/ssl_client_socket_openssl.cc (right):

https://codereview.chromium.org/1126113004/diff/1/net/socket/ssl_client_socke...
net/socket/ssl_client_socket_openssl.cc:943:
GotoState(STATE_HANDSHAKE_COMPLETE);
TBH, this transition is kinda pointless in the failing case, since
DoHandshakeComplete will just immediately return. But going from Foo to
FooComplete seems more consistent with the rest of net.

[Ryan Sleevi, 2015-05-08 22:49:17]

lgtm

[davidben, 2015-05-12 22:12:50]

The CQ bit was checked by davidben@chromium.org

[commit-bot: I haz the power, 2015-05-12 22:15:00]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1126113004/1

[commit-bot: I haz the power, 2015-05-12 22:30:23]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-05-12 22:31:22]

Patchset 1 (id:??) landed as
https://crrev.com/c4212c0b5cc733092b688b5eb1fa53eea1b15c08
Cr-Commit-Position: refs/heads/master@{#329518}