Add integration test for POD v4/v6 egress traffic

[wanyufe]

What type of PR is this?
feature - add integration testing for POD v4/v6 egress in v6/v4 clusters

Which issue does this PR fix:
N/A

What does this PR do / Why do we need it:
Increase integration test coverage for POD egress.

If an issue # is not available please add repro steps and logs from IPAMD/CNI showing the issue:

Testing done on this change:

Integration testing

~/amazon-vpc-cni-k8s/test/integration/cni-egress   egress-part-3 ✚ ● ? ⍟3  ginkgo -v --fail-on-pending -- \           1 ↵  408  18:18:30
 --cluster-kubeconfig=$KUBECONFIG \
 --cluster-name=$CLUSTER_NAME_V4 \
 --aws-region=$AWS_REGION \
 --aws-vpc-id=$VPC_ID_V4 \
 --ng-name-label-key=$NG_NAME_LABEL_KEY_V4 \
 --ng-name-label-val=$NG_NAME_LABEL_VAL_V4

Running Suite: CNI Pod Networking Suite - ~/amazon-vpc-cni-k8s/test/integration/cni-egress
======================================================================================================
Random Seed: 1683325258

Will run 1 of 1 specs
------------------------------
[BeforeSuite] 
~/amazon-vpc-cni-k8s/test/integration/cni-egress/pod_egress_suite_test.go:51
  STEP: checking cluster v4 or v6 @ 05/05/23 18:21:09.943
  STEP: creating test namespace @ 05/05/23 18:21:10.141
  STEP: getting the node with the node label key eks.amazonaws.com/nodegroup and value ipv4-nodegroup @ 05/05/23 18:21:10.164
  STEP: verifying at least 1 node present for the test @ 05/05/23 18:21:10.265
  STEP: getting the instance type from node label beta.kubernetes.io/instance-type @ 05/05/23 18:21:10.426
  STEP: getting the network interface details from ec2 @ 05/05/23 18:21:10.426
  STEP: setting the environment variables on the ds to map[ENABLE_V6_EGRESS:true] @ 05/05/23 18:21:10.465
  STEP: getting the aws-node daemon set in namespace kube-system @ 05/05/23 18:21:10.465
  STEP: updating the daemon set with new environment variable @ 05/05/23 18:21:10.566
  STEP: setting the environment variables on the ds to map[ENABLE_V6_EGRESS:true] @ 05/05/23 18:21:18.607
  STEP: getting the aws-node daemon set in namespace kube-system @ 05/05/23 18:21:18.607
  STEP: updating the daemon set with new environment variable @ 05/05/23 18:21:18.608
[BeforeSuite] PASSED [19.020 seconds]
------------------------------
[CANARY] test cluster egress connectivity container can access off-cluster service using egress interface
~/amazon-vpc-cni-k8s/test/integration/cni-egress/pod_egress_test.go:131
  STEP: creating test deployment on primary node: ip-192-168-34-162.us-east-2.compute.internal @ 05/05/23 18:21:28.65
  STEP: fetching pod primarynode-egress-tester-6b89dd6fd9-nqhxp egress address ... @ 05/05/23 18:21:50.997
  STEP: fetching pod primarynode-egress-tester-6b89dd6fd9-bf4jr egress address ... @ 05/05/23 18:21:51.161
  STEP: fetching pod primarynode-egress-tester-6b89dd6fd9-9xgf8 egress address ... @ 05/05/23 18:21:51.298
  STEP: fetching pod primarynode-egress-tester-6b89dd6fd9-vdpkt egress address ... @ 05/05/23 18:21:51.451
  STEP: testing pods in primary ENI IPv6 egress running in primary node: ip-192-168-34-162.us-east-2.compute.internal @ 05/05/23 18:21:51.594
  STEP: testing pods in secondary ENI IPv6 egress running in primary node: ip-192-168-34-162.us-east-2.compute.internal @ 05/05/23 18:21:53.531
  STEP: testing IPv6 ping between PODs within same ENI using egress interface is blocked @ 05/05/23 18:21:56.055
  STEP: testing IPv6 ping between PODs within different ENIs using egress interface is blocked @ 05/05/23 18:23:42.813
  STEP: deleting the primary node egress-tester deployment @ 05/05/23 18:26:09.59
• [280.961 seconds]
------------------------------
[AfterSuite] 
~/amazon-vpc-cni-k8s/test/integration/cni-egress/pod_egress_suite_test.go:117
  STEP: deleting test namespace @ 05/05/23 18:26:09.611
  STEP: reset daemonset back @ 05/05/23 18:26:19.731
  STEP: removing the environment variables from the ds map[ENABLE_V6_EGRESS:{}] @ 05/05/23 18:26:19.731
  STEP: getting the aws-node daemon set in namespace kube-system @ 05/05/23 18:26:19.731
  STEP: updating the daemon set with new environment variable @ 05/05/23 18:26:19.731
  STEP: update environment variables map[AWS_VPC_ENI_MTU:9001 AWS_VPC_K8S_CNI_VETHPREFIX:eni], remove map[ENABLE_V6_EGRESS:{} WARM_ENI_TARGET:{} WARM_IP_TARGET:{}] @ 05/05/23 18:26:27.775
  STEP: getting the aws-node daemon set in namespace kube-system @ 05/05/23 18:26:27.775
  STEP: updating the daemon set with new environment variable @ 05/05/23 18:26:27.775
[AfterSuite] PASSED [30.282 seconds]
------------------------------

Ran 1 of 1 Specs in 330.264 seconds
SUCCESS! -- 1 Passed | 0 Failed | 0 Pending | 0 Skipped
PASS | FOCUSED

Ginkgo ran 1 suite in 5m41.369616228s
Test Suite Passed
Detected Programmatic Focus - setting exit status to 197

Automation added to e2e:

Will this PR introduce any new dependencies?:

No
Will this break upgrades or downgrades. Has updating a running cluster been tested?:
No

Does this change require updates to the CNI daemonset config files to work?:

No
Does this PR introduce any user-facing change?:

No

This commit requires v4 cluster provisioned by Prow has ipv6 address assigned to node primary ENI.

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

[jdn5126]

Test overall looks good, just some comments on simplification

[wanyufe]

The code was refactored based on above comments and further discussion with @jdn5126. Here is the summary of changes:

1. test specs were moved to a new test suite (folder named cni-egress) specially purposed for egress connectivity testing.
   The reason behind is that for each test suite, PROW will provision a proper cluster to run that suite, and egress testing for IPv4 cluster has a non-default cluster env (ENABLE_V6_EGRESS = 'true'), the cluster provisioned for original cni test suite has not that env set to true.
2. code related to secondary node had been removed - test spec only runs on primary node. Team agreed that blocking testing between pods cross primary and secondary nodes is redundant.
3. Not to be confused with No. 2, blocking testing among pods cross primary ENI and secondary ENI is kept - both primary ENI and secondary ENI assigned to the same primary node.

[jdn5126]

Few nits, otherwise this looks good to go