http: Optimize queued client aborts

[kpdecker]

Avoid sending unsent data and destroying otherwise legitimate sockets
for requests that are aborted while still in the agent queue. This
reduces stress on upstream systems who will likely respond to the
request but client app already knows that it will be dropped on the
floor and also helps avoid killing keep-alive connections.

[Nodejs-Jenkins]

Thank you for contributing this pull request! Here are a few pointers to make sure your submission will be considered for inclusion.

The following commiters were not found in the CLA:

• kpdecker

You can fix all these things without opening another issue.

Please see CONTRIBUTING.md for more information

[indutny]

Are you sure that there is no chance of the data coming out before .abort()?

[kpdecker]

This only changes the behavior of abort if there is not a socket. Since there is no socket, no data could have been sent unless I'm missing something.

[trevnorris]

Two things.

1. Please add ClientRequest.prototype.aborted = false; just below the function ClientRequest() declaration to set the default value.

2. Please include a test for this change in functionality.

Ping me when this is done. :)

[kpdecker]

@trevnorris Done.

Rebuilding after switching between 0.10 and master apparently takes a bit of time :)

[trevnorris]

Thanks much.

@tjfontaine This LGTM, but mind giving it a glance to make sure there was nothing I missed?

[tjfontaine]

Can we make this by default undefined and then when set Date.now()? Seems like it would be helpful information when debugging (wallclock usage aside)

[kpdecker]

The aborted flag?

[tjfontaine]

@kpdecker yup

[kpdecker]

@tjfontaine Updated

[tjfontaine]

Thanks, squashed and landed in d0c7d93