Go: use package grouping in existing models-as-data models

[owen-mc]

Use the mechanism introduced in #16941 in the existing models.

[github-actions]

⚠️ The head of this PR and the base branch were compared for differences in the framework coverage reports. The generated reports are available in the artifacts of this workflow run. The differences will be picked up by the nightly job after the PR gets merged.

Click to show differences in coverage

go

Generated file changes for go

• Changes to framework-coverage-go.rst:

-    `Go JOSE <https://github.com/go-jose/go-jose>`_,"``github.com/go-jose/go-jose*``, ``github.com/square/go-jose*``, ``gopkg.in/square/go-jose*``",,8,2
+    `Go JOSE <https://github.com/go-jose/go-jose>`_,"``github.com/go-jose/go-jose*``, ``github.com/square/go-jose*``, ``gopkg.in/square/go-jose*``",,12,3
-    Totals,,204,898,24
+    Totals,,204,902,25

• Changes to framework-coverage-go.csv:

+ github.com/square/go-jose/jwt,1,,4,,1,,4,
- gopkg.in/square/go-jose.v2/jwt,1,,4,,1,,4,
+ gopkg.in/square/go-jose/jwt,1,,4,,1,,4,

[mbg]

Looks good. I am wondering about the (need for) the last commit though. For the purpose of those tests, does it really matter which line they come from or could the name of the file or some other, constant information suffice?

[owen-mc]

I'm not sure I totally follow. The last commit doesn't relate to line numbers. Currently, many tests include provenance information for edges which uses a different number for each models-as-data model. Adding or removing models perturbs these numbers, and causes a lot of unrelated test changes. There are plans in motion to fix this soon, but in the meantime we are just having to work around the constant unrelated test churn that makes all models-related tests conflict with each other.

[mbg]

Sorry, I had just assumed that they were line numbers rather than something else. That doesn't change my question though, but you have answered it anyway that there are plans to change this.