Move scm information to metadata from top-level, and remove nesting

[holly-cummins]

Follow-up to change the design of #28009 a bit. Adding scm information at the top level needs schema changes in the registry, and the information should logically be considered metadata. I've moved it down a level, and got rid of nesting to make sure tools don't get confused by nested structures within the nested structure.

[quarkus-bot]

Failing Jobs - Building 1a7563f

Status	Name	Step	Failures	Logs	Raw logs
✔️	Maven Tests - JDK 11
⌛	Maven Tests - JDK 11 Windows	Build	⚠️ Check →	Logs	Raw logs
✖	Native Tests - Messaging1	Build	Failures	Logs	Raw logs

Full information is available in the Build summary check run.

Failures

⚙️ Native Tests - Messaging1 #

- Failing: integration-tests/reactive-messaging-kafka 

📦 integration-tests/reactive-messaging-kafka

✖ io.quarkus.it.kafka.KafkaConnectorIT.testPets - More details - Source on GitHub

java.lang.RuntimeException: java.lang.RuntimeException: Unable to successfully launch process '21987'. Exit code is: '1'.
	at io.quarkus.test.junit.QuarkusIntegrationTestExtension.throwBootFailureException(QuarkusIntegrationTestExtension.java:337)
	at io.quarkus.test.junit.QuarkusIntegrationTestExtension.beforeEach(QuarkusIntegrationTestExtension.java:110)

[holly-cummins]

I think the failing kafka tests must be unrelated to these changes. The maven windows tests are more suspicious, but looking through the logs, I can't see anything obvious. Does that build sometimes run out of time?