refactor(serving): KnClient interface for single point of cluster access [ginkgo version]

[rhuss]

This PR exactly the same as #134 except that the test is implemented with Gingko.

The purpose of this PR is to get a feeling for the difference in writing tests in Ginkgo and/or a testframework baset on Golang's intrinsic testing with some sugar added.

The relevant tests to compare are:

• Ginkgo --> https://github.com/rhuss/client/blob/pr/add-schema-handling-ginko-test/pkg/serving/v1alpha1/client_test.go
• gotest.tools --> https://github.com/rhuss/client/blob/pr/add-schema-handling/pkg/serving/v1alpha1/client_test.go

Follow the discussion overthere whether to pick up this PR or #134

[knative-prow-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: rhuss

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 [rhuss]

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

[knative-metrics-robot]

The following is the coverage report on pkg/.
Say /test pull-knative-client-go-coverage to re-run this coverage report

File	Old Coverage	New Coverage	Delta
pkg/kn/commands/namespaced.go	87.5%	72.7%	-14.8
pkg/kn/commands/revision/revision_describe.go	79.2%	78.3%	-0.9
pkg/kn/commands/revision/revision_list.go	81.2%	80.6%	-0.7
pkg/kn/commands/service/service_create.go	69.0%	68.6%	-0.4
pkg/kn/commands/service/service_describe.go	79.2%	78.3%	-0.9
pkg/kn/commands/service/service_list.go	80.0%	79.2%	-0.8
pkg/serving/schema_handling.go	Do not exist	91.7%
pkg/serving/v1alpha1/client.go	Do not exist	75.7%

[maximilien]

Thanks for doing this work @rhuss!
Nice of you going the extra mile to show concrete non-trivial example.
I am ambivalent as to which wins. The great thing is either will be better than what we have now.
So happy whichever decision we (as a team makes): Ginkgo or gotest.tools.

🍻

[rhuss]

I am ambivalent as to which wins. The great thing is either will be better than what we have now.
So happy whichever decision we (as a team makes): Ginkgo or gotest.tools.

cool, thanks for your feedback. Let's see how this discussion goes tomorrow, but I expect a definitive decision then.

Then, when we merge either of the two PR, we have also the benefit that in the command unit tests we soon don't have to bother with faking the knative API directly (this is done in the test for KnClient as shown), but with an own mock implementation of the KnClient interface. I will work on this soon after one of these PRs are merged.

[rhuss]

closing as the gotest.tools version from #134 goes in.