ci: use registry cache backend instead of gha

[rjaegers]

Pull Request

Description of changes

This PR moves the build-push Docker cache from the GitHub Actions (gha) cache back-end to the registry back-end. The cached results will be pushed to a (private) package in the registry and used for subsequent builds. Contrary to the gha cache this does not take ancestry into account. All branches can freely use this cache, hence the removal of the Docker step in prime-cache.

Some more details about the different caches can be found here: https://blacksmith.sh/blog/cache-is-king-a-guide-for-docker-layer-caching-in-github-actions.

Checklist

• I have followed the contribution guidelines for this repository
• I have added tests for new behavior, and have not broken any existing tests
• I have added or updated relevant documentation
• I have verified that all added components are accounted for in the SBOM

[github-actions]

🦙 MegaLinter status: ✅ SUCCESS

Descriptor	Linter	Files	Fixed	Errors	Elapsed time
✅ ACTION	actionlint	14		0	0.06s
✅ DOCKERFILE	hadolint	2		0	0.52s
✅ JSON	prettier	11	0	0	0.92s
✅ JSON	v8r	10		0	4.48s
✅ MARKDOWN	markdownlint	8	0	0	1.1s
✅ MARKDOWN	markdown-table-formatter	8	0	0	0.25s
✅ REPOSITORY	checkov	yes		no	18.27s
✅ REPOSITORY	gitleaks	yes		no	0.43s
✅ REPOSITORY	git_diff	yes		no	0.01s
✅ REPOSITORY	grype	yes		no	13.91s
✅ REPOSITORY	secretlint	yes		no	1.69s
✅ REPOSITORY	trivy	yes		no	7.15s
✅ REPOSITORY	trivy-sbom	yes		no	1.63s
✅ REPOSITORY	trufflehog	yes		no	3.53s
✅ SPELL	lychee	48		0	1.42s
✅ YAML	prettier	19	0	0	1.31s
✅ YAML	v8r	19		0	14.81s
✅ YAML	yamllint	19		0	0.52s

See detailed report in MegaLinter reports

MegaLinter is graciously provided by

[github-actions]

Test Results

 2 files  ±0   2 suites  ±0   51s ⏱️ ±0s
26 tests ±0  26 ✅ ±0  0 💤 ±0  0 ❌ ±0 
28 runs  ±0  28 ✅ ±0  0 💤 ±0  0 ❌ ±0 

Results for commit d846b39. ± Comparison against base commit 24442b0.

♻️ This comment has been updated with latest results.

[github-actions]

Compressed layer size comparison

Comparing ghcr.io/philips-software/amp-devcontainer-rust:latest to ghcr.io/philips-software/amp-devcontainer-rust@sha256:5bfcb8b58a1991fe77291bf8e806be3a70ebfb8e008be526ac313eb8c0ca259b

OS/Platform	Previous Size	Current Size	Delta
linux/amd64	432.86M	432.86M	188.00 (+0.00%)
linux/arm64	572.09M	572.09M	-468.00 (-0.00%)

[github-actions]

Compressed layer size comparison

Comparing ghcr.io/philips-software/amp-devcontainer-cpp:latest to ghcr.io/philips-software/amp-devcontainer-cpp@sha256:d60fad6aa6cf489c7a0d42b1fe925622c2547752a5394ec89594122bf9c55094

OS/Platform	Previous Size	Current Size	Delta
linux/amd64	640.27M	640.27M	-217.00 (-0.00%)
linux/arm64	632.60M	632.62M	15.72K (+0.00%)

[daantimmer]

Care to explain the reason behind this change?

[rjaegers]

It is/was an experiment to see if the cache works better when using a different back-end.

There are some pros/cons to this approach. Some of which can be found here: https://blacksmith.sh/blog/cache-is-king-a-guide-for-docker-layer-caching-in-github-actions.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
No data about Coverage
No data about Duplication

See analysis details on SonarCloud

[rjaegers]

Closing this one as it does not seem to provide benefits over GHA approach.

[github-actions]

Pull Request Report (#443)

Static measures

Description	Value
Number of added lines	10
Number of deleted lines	18
Number of changed files	4
Number of commits	8
Number of reviews	1
Number of comments (w/o review comments)	8
Number of reviews that contains a comment to resolve	0
Number of reviews that requested a change from the author	0
Number of reviews that approved the Pull Request	1
Get the total number of participants of a Pull Request	6

Time related measures

Description	Value
PR lead time (from creation to close of PR)	1.8 Days
Time that was spend on the branch before the PR was created	4.6 Min
Time that was spend on the branch before the PR was merged	0 Sec
Time to merge after last review	0 Sec

Status check related measures

Description	Value
Total runtime for last status check run (Workflow for PR)	12.6 Min
Total time spend in last status check run on PR	8.4 Min