Remove tf-operator from the codebase (kubeflow#1378)

* Remove cmd/tf-operator

With training-operator.v1 in place, we no longer need tf-operator.v1

* Move util_test.go to v1/testutil

The tests were effectively testing the contract of the
testutil package so it makes sense the tests exist
within the package itself.

* Remove TFController code and tests

With TFReconciler in place ce can remove the TFController and
associated tests.

* Remove tf_operator build dockerfile

With tf_operator code deleted, we no longer need the dockerfile
to build an image to run tf_operator.

* Update executable name tf_operator/release.py

Substitute the binary from tf-operator to training-operator.

* Update the developer_guide.md to use the training-operator

- Substitute references to tf-operator with training-operator
- Add instructions to apply all job CRDs instead of just TFJob as the
  operator expects them all to be present.

* Use make commands in developer guide

replace kubectl apply commands with makefile targets.

* Remove stale metric counters

Metric counters have been refactored into pkg/common/metrics.go. This
diff removes the stale counters present in tfjobcontroller.

* Remove util_test.go

As per review comments, we don't need to move
the util_tests.go in pkg/common/util/testutil/util_test.go

* Updated comments on the moved helper functions

Add remark about the origin file for the helper methods that
were moved during the refactor in pkg/tensorflow.

* Remove unused variables

delete variables no longer in use.

* Set controller name from tf-operator to tfjob-operator

tfjob-operator is more consistent with the naming of other operators
such as pytorchjob-operator or xgboostjob-operator

* Rename controller names to xxjob-controller

- xgboostjob-operator -> xgboostjob-controller
- mxnet-operator -> mxjob-controller
- pytorchjob-operator -> pytorchjob-controller
- tfjob-operator -> tfjob-controller