fix: clean node_modules before install in Inferno CI

[quotentiroler]

Fixes esbuild version mismatch by clearing node_modules before install

[Copilot]

Pull request overview

This PR addresses an esbuild version mismatch issue in the Inferno CI workflow by clearing node_modules before installation. The changes include:

• Adding rm -rf node_modules */node_modules before bun install to ensure clean dependency installation
• Migrating from Docker Compose-based Inferno setup to direct Ruby execution
• Adding service containers (Inferno DB, Redis) to the GitHub Actions workflow
• Enhancing error handling and logging throughout the test execution
• Updating package versions across multiple package.json files

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
.github/workflows/smart-compliance-tests.yml	Adds node_modules cleanup before install, migrates Inferno from Docker to direct Ruby execution, adds service containers, and improves error handling
ui/package.json	Updates version string to include new build timestamp
scripts/package.json	Updates version string to include new build timestamp
package.json	Updates version string to include new build timestamp
backend/package.json	Updates version string to include new build timestamp

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The wait time for Keycloak has been increased from 60 to 90 iterations (7.5 minutes total). This increase should be documented with a comment explaining why the longer timeout is necessary, or consider if this indicates an underlying performance issue that should be addressed.

Suggested change

	echo "Waiting for Keycloak..."
	echo "Waiting for Keycloak..."
	# NOTE: This loop was intentionally increased to 90 iterations (7.5 minutes max)
	# to account for slow Keycloak startup in CI (e.g., cold image pulls, DB migrations).
	# If Keycloak frequently requires close to this full timeout, underlying performance
	# issues (container startup time, database readiness, etc.) should be investigated.

[Copilot]

The emoji indicators (✅, ❌, ⏭️, ⚠️) were removed from the status column. While this change makes the output cleaner, it reduces visual scanning efficiency. Consider restoring the emojis or document why they were removed.

Suggested change

	echo "| Passed | $PASSED |" >> $GITHUB_STEP_SUMMARY
	echo "| Failed | $FAILED |" >> $GITHUB_STEP_SUMMARY
	echo "| Skipped | $SKIPPED |" >> $GITHUB_STEP_SUMMARY
	echo "| Errors | $ERRORS |" >> $GITHUB_STEP_SUMMARY
	echo "| ✅ Passed | $PASSED |" >> $GITHUB_STEP_SUMMARY
	echo "| ❌ Failed | $FAILED |" >> $GITHUB_STEP_SUMMARY
	echo "| ⏭️ Skipped | $SKIPPED |" >> $GITHUB_STEP_SUMMARY
	echo "| ⚠️ Errors | $ERRORS |" >> $GITHUB_STEP_SUMMARY

[Copilot]

Adding continue-on-error: true silently allows test failures to pass the CI check. This undermines the purpose of compliance testing. If failures should be non-blocking during development, this should be controlled via a workflow input or environment variable rather than being hardcoded.