Ref - https://github.com/appscode/g2/issues/28

* fix worker reconnect errors, send cando after reconnection.

Fix race condition in client.Do()

Fixes duplicated code after merge

OS and configuration specific integration test

Run integration tests with the -integration flag

fixed typo on WorkComplete

Add lock on job handler assignment

Fixes race condition on jobs being done before handler is set.

Add sync lock to create job functions

Add sync lock to make create job calls thread safe.

concurrent map bug fixed #70

fixed #65

Allow built in selectionHandlers to be reused outside of package

Export SelectionHandler

Replace mutex in client.do() with a channel to avoid deadlock and int…

Updated to reflect new type names.

Fixed WORK_FAIL needing 2 arguments

According to http://gearman.org/protocol/ , WORK_FAIL only has 1 argument: the handle

Do not ignore write errors

Avoid read channel corruption when response size > bufferSize

When receiving a response, what was happening

1. Read bufferSize and it gets assigned to leftdata
2. Read another bufferSize
3. 2 buffers get appended, but leftdata still points to first buffer
4. Process data buffer which contains only complete responses
5. Back to ReadLoop, but leftdata still points to first incomplete buffer
    causing corrupt data to be processed

Solution is to make leftdata nil once we have merged it with the second buffer

When a job completes, remove its handler