Fix HttpStreamParser fuzzer (again)

Fix HttpStreamParser fuzzer.

It was failing to read the body.  0 while reading the headers means OK,
while 0 while reading the body means the connection was closed. I fixed
this issue two weeks ago, but then re-introduced it in a refactor.

I think that shows how weird and confusing network stack result handling
is.  :(

BUG=598037
Committed: https://crrev.com/d284559ad772c498d5dae1a88d11a041f2941ffe
Cr-Commit-Position: refs/heads/master@{#389314}

[mmenke, 2016-04-22 20:57:56]

mmenke@chromium.org changed reviewers:
+ eroman@chromium.org

[mmenke, 2016-04-22 20:57:57]

*grumble*....

[eroman, 2016-04-22 22:13:00]

you mean there are problems from overloading the meaning of a single signed
integer between meaning error, success, or in other cases non-negative number of
bytes, sometimes with 0 having special meaning success and other times EOF?
without having type enforcement by the compiler?

inconceivable!

</snark>

[eroman, 2016-04-22 22:13:20]

lgtm

[mmenke, 2016-04-22 22:51:33]

The CQ bit was checked by mmenke@chromium.org

[commit-bot: I haz the power, 2016-04-22 22:52:01]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1920463003/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1920463003/1

[commit-bot: I haz the power, 2016-04-23 00:01:00]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2016-04-23 00:03:25]

Description was changed from

==========
Fix HttpStreamParser fuzzer.

It was failing to read the body.  0 while reading the headers means OK,
while 0 while reading the body means the connection was closed. I fixed
this issue two weeks ago, but then re-introduced it in a refactor.

I think that shows how weird and confusing network stack result handling
is.  :(

BUG=598037
==========

to

==========
Fix HttpStreamParser fuzzer.

It was failing to read the body.  0 while reading the headers means OK,
while 0 while reading the body means the connection was closed. I fixed
this issue two weeks ago, but then re-introduced it in a refactor.

I think that shows how weird and confusing network stack result handling
is.  :(

BUG=598037

Committed: https://crrev.com/d284559ad772c498d5dae1a88d11a041f2941ffe
Cr-Commit-Position: refs/heads/master@{#389314}
==========

[commit-bot: I haz the power, 2016-04-23 00:03:26]

Patchset 1 (id:??) landed as
https://crrev.com/d284559ad772c498d5dae1a88d11a041f2941ffe
Cr-Commit-Position: refs/heads/master@{#389314}