Migrate v1 implementation to latest kubeflow/common

[Jeffwan]

Address #83 and part of #44

As you can see in the PR,

1. We don't need to implement Create/Delete Pod/Service. As this is pretty generic, kubeflow/common already provides implementation
2. Batch scheduler has been changed from kube-batch to volcano. By default, it's disabled in XGBoost operator. This is just reversed for users who need it.
3. expectation.go starts to use expectation package from kubeflow/common.
4. package changes from

• "github.com/kubeflow/common/job_controller/api/v1"
• "github.com/kubeflow/common/job_controller"
  to
• "github.com/kubeflow/common/pkg/apis/common/v1"
• "github.com/kubeflow/common/pkg/controller.v1/common"

We can easily get performance improvement, bug fixed via common version upgrade.

Kubeflow/common will be more and more stable, the changes will be few lines in the future.

[kubeflow-bot]

This change is 

[terrytangyuan]

Looks great! Thanks

/lgtm
/approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: terrytangyuan

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [terrytangyuan]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[merlintang]

Why do we require to use volcano ? in most of the cases, we do not have kubebatch or volcano. @Jeffwan

[merlintang]

for common, volcano is the default scheduler?

[Jeffwan]

@merlintang Yes. Most of the kube-batch contributors come to volcano project now. This is a kind of placeholder and make sure it's common compatible. In the previous implementation, we import kube-batch as well but not use it.

xgboost itself can still be installed and used separately. This is transparent to users and user don't need to use it. I should add more comments on the code.

One improvement we can do here is to hide these details to different operators. I will add one issue kubeflow/common#96