test/extended/cli/adm_upgrade/recommend: Allow Failing!=False

A depressing amount of the time in single-node testing, the
authentication ClusterOperator is Available=False, and causes the
'recommend' command to error if the 'Failing' risk is not accepted.
For example, in this aggregation run's 20 attempts [1]:

* 9 passed.
* three failed the '--version' test-case on 'authentication' 'Available=False' [2,3,4].
* two failed on build-cluster registry 500s [5,6].
* two failed on 'prometheus-operator' watch requests [7,8].
* one failed on a 'context deadline exceeded' out of
  'runUpdateService' [9], with 'Failed to pull image...authentication
  required' issues trying to get the 'tools' image.
* one failed on authentication Pod restarts [10].
* one failed on an un-excepted 'authentication' 'Available=False'
  ('OAuthServerDeployment_NoPod') [11].
* one failed on an unexpected successful return in an
  TestImageStreamTagsAdmission test-case [12].

So the auth functionality is pretty flaky in these single-node
clusters under serial-suite load.  Ideally [13] can get addressed or
the auth component can otherwise get firmed up, but until then, this
commit softens our logic to allow that kind of ClusterOperator
Available=False (which gets bubbled up as ClusterVersion
Failing=True), without failing our new test-case.

[1]: #30113 (comment)
[2]: https://prow.ci.openshift.org/view/gs/test-platform-results/logs/openshift-origin-30113-nightly-4.21-e2e-aws-ovn-single-node-serial/1964055948225941504
[3]: https://prow.ci.openshift.org/view/gs/test-platform-results/logs/openshift-origin-30113-nightly-4.21-e2e-aws-ovn-single-node-serial/1964055951426195456
[4]: https://prow.ci.openshift.org/view/gs/test-platform-results/logs/openshift-origin-30113-nightly-4.21-e2e-aws-ovn-single-node-serial/1964055945973600256
[5]: https://prow.ci.openshift.org/view/gs/test-platform-results/logs/openshift-origin-30113-nightly-4.21-e2e-aws-ovn-single-node-serial/1964055945063436288#1:build-log.txt%3A32
[6]: https://prow.ci.openshift.org/view/gs/test-platform-results/logs/openshift-origin-30113-nightly-4.21-e2e-aws-ovn-single-node-serial/1964055945512226816#1:build-log.txt%3A43
[7]: https://prow.ci.openshift.org/view/gs/test-platform-results/logs/openshift-origin-30113-nightly-4.21-e2e-aws-ovn-single-node-serial/1964055950948044800
[8]: https://prow.ci.openshift.org/view/gs/test-platform-results/logs/openshift-origin-30113-nightly-4.21-e2e-aws-ovn-single-node-serial/1964055948683120640
[9]: https://prow.ci.openshift.org/view/gs/test-platform-results/logs/openshift-origin-30113-nightly-4.21-e2e-aws-ovn-single-node-serial/1964055952327970816
[10]: https://prow.ci.openshift.org/view/gs/test-platform-results/logs/openshift-origin-30113-nightly-4.21-e2e-aws-ovn-single-node-serial/1964055953233940480
[11]: https://prow.ci.openshift.org/view/gs/test-platform-results/logs/openshift-origin-30113-nightly-4.21-e2e-aws-ovn-single-node-serial/1964055947772956672
[12]: https://prow.ci.openshift.org/view/gs/test-platform-results/logs/openshift-origin-30113-nightly-4.21-e2e-aws-ovn-single-node-serial/1964055952797732864
[13]: https://issues.redhat.com/browse/OCPBUGS-20056

Author: wking

test/extended/cli/adm_upgrade/recommend.go

@@ -240,7 +240,7 @@ Updates to 4[.][0-9]*:
 				o.Expect(oc.Run("config", "set-credentials").Args("test", "--token", token).Execute()).To(o.Succeed())
 				o.Expect(oc.Run("config", "set-context").Args("--current", "--user", "test").Execute()).To(o.Succeed())
 
-				out, err := oc.Run("--certificate-authority", caBundleFilePath, "adm", "upgrade", "recommend", "--version", fmt.Sprintf("4.%d.0", currentVersion.Minor+1), "--accept", "ConditionalUpdateRisk").EnvVar("OC_ENABLE_CMD_UPGRADE_RECOMMEND", "true").EnvVar("OC_ENABLE_CMD_UPGRADE_RECOMMEND_PRECHECK", "true").EnvVar("OC_ENABLE_CMD_UPGRADE_RECOMMEND_ACCEPT", "true").Output()
+				out, err := oc.Run("--certificate-authority", caBundleFilePath, "adm", "upgrade", "recommend", "--version", fmt.Sprintf("4.%d.0", currentVersion.Minor+1), "--accept", "ConditionalUpdateRisk,Failing").EnvVar("OC_ENABLE_CMD_UPGRADE_RECOMMEND", "true").EnvVar("OC_ENABLE_CMD_UPGRADE_RECOMMEND_PRECHECK", "true").EnvVar("OC_ENABLE_CMD_UPGRADE_RECOMMEND_ACCEPT", "true").Output()
 
 				o.Expect(err).NotTo(o.HaveOccurred())
 				err = matchRegexp(out, `The following conditions found no cause for concern in updating this cluster to later releases.*
@@ -253,7 +253,7 @@ Image: example.com/test@sha256:ccccccccccccccccccccccccccccccccccccccccccccccccc
 Release URL: https://example.com/release/4[.][0-9]*[.]0
 Reason: accepted (TestRiskA|MultipleReasons) via ConditionalUpdateRisk
 Message: (?s:.*)This is a test risk[.] https://example.com/testRiskA
-Update to 4[.][0-9]*[.]0 has no known issues relevant to this cluster other than the accepted ConditionalUpdateRisk.`)
+Update to 4[.][0-9]*[.]0 has no known issues relevant to this cluster other than the accepted ConditionalUpdateRisk(|,Failing).`)
 				o.Expect(err).NotTo(o.HaveOccurred())
 			})
 		})