Remove client_auth_cert_needed_ from SSLClientSocketOpenSSL.

Remove client_auth_cert_needed_ from SSLClientSocketOpenSSL.

That state isn't needed. When an operation fails in OpenSSL, SSL_get_error
tells you exactly what event it was waiting for.

BUG=none

Committed: https://crrev.com/ced4aa9b28c59dac38971c3528f3bcc05d700dde
Cr-Commit-Position: refs/heads/master@{#329491}

[davidben, 2015-05-08 22:15:52]

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

[davidben, 2015-05-08 22:15:52]

To verify BoringSSL's semantics, grep for SSL_X509_LOOKUP. SSL_get_error is this
monster of a function and -1 vs 0 are suuuuuuuper mixed up. But the first-order
approximation is that you return SSL_ERROR_WANT_FOO if rwstate told you that's
when you failed, unless there's an error in the error queue. Then it returns
SSL_ERROR_SSL.

https://codereview.chromium.org/1136813002/diff/20001/net/socket/ssl_client_s...
File net/socket/ssl_client_socket_openssl.cc (right):

https://codereview.chromium.org/1136813002/diff/20001/net/socket/ssl_client_s...
net/socket/ssl_client_socket_openssl.cc:1403: !ssl_config_.send_client_cert) {
This is the same as checking it before the ssl_ret check. If
client_auth_cert_needed_ was true, we just returned -1 from a callback, which
means that ssl_ret is in a failing state. This is just more correct w.r.t. how
one is supposed to do OpenSSL error-handling.

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

(This sites on top of the renego CL, so I haven't fired off try jobs yet.)

[Ryan Sleevi, 2015-05-08 22:52:38]

lgtm

[davidben, 2015-05-12 20:58:02]

The CQ bit was checked by davidben@chromium.org

[commit-bot: I haz the power, 2015-05-12 20:58:45]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1136813002/20001

[commit-bot: I haz the power, 2015-05-12 21:22:47]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-05-12 21:23:39]

Patchset 2 (id:??) landed as
https://crrev.com/ced4aa9b28c59dac38971c3528f3bcc05d700dde
Cr-Commit-Position: refs/heads/master@{#329491}