[core] Fix lazy init for allSuppressors in DefaultRuleViolationFactory

[adangel]

Describe the PR

I've decided not to use AtomicReference as suggested but instead use a simpler approach with volatile. Both solution wouldn't avoid creating the HashSet more than once (for that we would need to synchronize completely), but that shouldn't hurt, since the result should always be the same (always the same suppressors in the Set).

Related issues

• Fixes [core] ConcurrentModificationException in DefaultRuleViolationFactory #4035

Ready?

• Added unit tests for fixed bug/feature
• Passing all unit tests
• Complete build ./mvnw clean verify passes (checked automatically by github actions)
• Added (in-code) documentation (if needed)

[pmd-test]

	2 Messages
📖	Compared to pmd/7.0.x: This changeset changes 10 violations, introduces 4 new violations, 0 new errors and 0 new configuration errors, removes 4 violations, 0 errors and 0 configuration errors. Full report
📖	Compared to master: This changeset changes 0 violations, introduces 699659 new violations, 22 new errors and 0 new configuration errors, removes 826993 violations, 39 errors and 6 configuration errors. Full report

Generated by 🚫 Danger

[oowekyala]

Let's merge this. Notably, #4050 deletes this logic, but I think the problem is not there in the new implementation (ref https://github.com/pmd/pmd/pull/4050/files#diff-2d8622e9c0e2553c0669b29d50f28d891fb80e348ec02fc078919dfc001dfb7aR162).