Fix test flake preventing BOM release

[basil]

This test was flaky, as it passed locally and passed in jenkinsci/bom#3137 but has failed twice on the main branch (e.g. https://ci.jenkins.io/job/Tools/job/bom/job/master/2712/testReport/) with

java.lang.AssertionError: 

Expected: is "Get contextual object from internal APIs"
     but: was "This stage will be skipped."
	at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:20)
	at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:6)
	at io.jenkins.plugins.pipelinegraphview.utils.PipelineStepApiLegacyTest.getAllStepsReturnsStepsForComplexParallelBranches(PipelineStepApiLegacyTest.java:217)
	at java.base/java.lang.reflect.Method.invoke(Method.java:580)
	at org.jvnet.hudson.test.JenkinsRule$1.evaluate(JenkinsRule.java:656)
	at java.base/java.util.concurrent.FutureTask.run(FutureTask.java:317)
	at java.base/java.lang.Thread.run(Thread.java:1583)

Looking into this further, it seems that PCT builds on the main branch have defined BRANCH_NAME=master as an environment variable and this causes the test to fail. I could reproduce this locally with export BRANCH_NAME=master. The test seemed to be trying to get a stage to be skipped, but when running on master the stage wasn't skipped and the test started failing. By changing the test to not rely on master I could get the test to pass regardless of what that environment variable was defined to.

Testing done

As described above

[basil]

If perchance anyone happens to be awake in the next few hours and interested in helping, it would be great to take advantage of time zones to merge and release the above and restart a build of https://ci.jenkins.io/job/Tools/job/bom/job/master/ which will take several hours.

[basil]

Specifically the test was relying on the branch being named something other than master, which was true for https://github.com/jenkinsci/pipeline-graph-view-plugin where the branch is named main and was also true for jenkinsci/bom#3137 where the branch was named PR-something-or-other, but not true for https://github.com/jenkinsci/bom where the branch is actually named master

[alecharp]

This looks simple enough and the tests are passing.

[alecharp]

@gabrieleara do you think you could approve and merge this pull request so it can unblock the release process of the bom? Thanks.

[jglick]

🤔 https://github.com/jenkinsci/pipeline-model-definition-plugin/blob/41dd8ef6dd561452755641a2a3c76e69e9b2d0cb/pipeline-model-definition/src/main/resources/org/jenkinsci/plugins/pipeline/modeldefinition/when/impl/BranchConditionalScript.groovy#L39-L43 is run outside of a node block so it is picking up environment variables from the MasterComputer, which would be that of the controller JVM. Would be preferable for https://github.com/jenkinsci/pipeline-model-definition-plugin/blob/41dd8ef6dd561452755641a2a3c76e69e9b2d0cb/pipeline-model-definition/src/main/java/org/jenkinsci/plugins/pipeline/modeldefinition/when/impl/BranchConditional.java#L98 to look up https://github.com/jenkinsci/branch-api-plugin/blob/9d30c9c8450ef4dbd296c7232416284e949a498e/src/main/java/jenkins/branch/BranchNameContributor.java#L57-L65 directly (or more properly via https://javadoc.jenkins.io/plugin/scm-api/jenkins/scm/api/SCMHead.HeadByItem.html#findHead(hudson.model.Item) ), if anyone feels like a short project to make this more robust.