Add build to GH actions run

[christian-bromann]

Requirements

• Filling out the template is required. Any pull request that does not include enough information to be reviewed in a timely manner may be closed at the maintainers' discretion.
• All new code requires tests to ensure against regressions.

Description of the Change

This patch contains modifications to the karma.conf.js to include build ids to the tests. This will fix the invalid badge on the Readme which now shows unknown because no new builds have been created for a while.

Alternate Designs

None, build id is required to display correct badge information

Why should this be in core?

Because the readme lives in this repo.

Benefits

Showing up to date build status promotes MochaJS being a stable testing framework.

Possible Drawbacks

None.

Applicable issues

It is a documentation update.

[christian-bromann]

Also, in order to not show failing builds from PRs I recommend to create a user for main branch builds and use that one for everything merged into the main branch. Sauce Labs infrastructure unfortunately doesn't not yet know anything about your git environment.

[coveralls]

Coverage remained the same at 94.155% when pulling f4a0d41 on christian-bromann:cb-sauce-build into c6f874f on mochajs:master.

[juergba]

@christian-bromann thanks for this PR.
I have noticed the unknown label a few days ago, but didn't really know the cause.

[juergba]

@christian-bromann I pushed your branch to our repo in order to check our browser tests. They passed, but the ESLint tests failed.
I have fixed thoses linting errors.

@outsideris For PR's from forked repos the ESLint tests are not reliable since they pass inspite of linting errors. There is a job Tests / ESLint Results which is missing for forked repos. I suppose the reason is: pull-request events from forked repos don't have write access to our repo. Therefore the linting results (which is an artifact) are silently lost.

[juergba]

I'm going to merge this PR, I'm unsure wether it's going to work though.

Edit: the label still shows unknown, maybe the change will take some time.
Our coverall test has failed pretending that 94.155 < 94.155 (-5.8%).

[christian-bromann]

@juergba yeah, it seems that the image in the readme is cached by GitHub, adding a ?foo at the end makes it load fresh and it works now: