Upgrade to Docker 1.7.1 by default

Update code and docs to use - in flag names instead of _

export path in shippable and travis

kubelet: refactor SyncPods for better readability

Update travis and shippable to have less needless cruft

Remove BindClientConfigFlags entirely

Generate conversions/deep-copies for experimental.

Rescheduler design space doc.

Generate conversions/deep-copies for experimental. Currently we make (and register) duplicate functions but this is benign.

Add Birthcry event to kube-proxy

Remove redundant tests. We already check that no changes have been made in hack/verify-generated-*.sh.

Eventually we would like to replace the all-encompassing SyncPods function with
more well-defined, smaller functions. This would not only help with the
readability and profiling of the code, it'd also set in motion for the plans to
trigger pod worker individually based on the content of the pod updates.

This commit serves as the first step of that, while avoiding breaking all unit
tests by preserving the SyncPods function for the time being.

Updating swagger datafile as per latest swagger-ui code

fix ABAC tests

Strip out extra gunk from the end of an example output line.

Fix kube-push using gcloud compute instance-groups managed set-instance-template

Fix go vet errors

Currently it is commented out.

It looks like there are places where things were needed for one command
were being just needlessly copied. Which made the files harder to
recognize what mattered and what didn't...

Switch the order of delete, delete pods before deleting nodes.

They are unused.

Speedups for precommit hooks

Instead of calling rsync over and over and over and over and over and
over and over and over and over (and probably over) use one `cp`

Before:
real	0m5.247s
user	0m2.294s
sys	0m1.300s

After:
real	0m2.260s
user	0m2.230s
sys	0m0.936s

Right now some of the hack/* tools use `go run` and build almost every
time. There are some which expect you to have already run `go install`.
And in all cases the pre-commit hook, which runs a full build wouldn't
want to do either, since it just built!

This creates a new hack/after-build/ directory and has the scripts which
REQUIRE that the binary already be built. It doesn't test and complain.
It just fails miserably. Users should not be in this directory. Users
should just use hack/verify-* which will just do the build and then call
the "after-build" version. The pre-commit hook or anything which KNOWS
the binaries have been built can use the fast version.

Just to get everything in one place...

Hooks seems like they should be hooks. Not sure hack makes a lot more
sense, but it has more stuff already.

It isn't used anywhere, why have it?

We were setting etcd values, but we started etcd before we set them.
Thankfully we were just setting them to etcd defaults, so it all worked

Always use the --cluster-version flag for the gke provider.

Fix vagrant kube-up

Modify the TestCase TestServiceRegistryIPUpdate