fix(helmfile): Sync OpenProject resource limits with upstream helm chart

🪲 Bugfix

Sync OpenProject resource limits with upstream helm chart to avoid running out of memory quickly with OpenProject.

✅ Changes

We have seen multiple deployments where OpenProject ran out of memory when being accessed by 10-20 users simultaneously. When checking the upstream helm charts, we noticed that the upstream chart uses 4 GB, whereas we use 3 GB. This MR now syncs all upstream resource requests and limits with the upstream helm chart once.

🧪 How to reproduce & test

Steps to reproduce

1. Run a deployment with the default resource limits (resources.yaml.gotmpl).
2. Let 20+ users OpenProject simultaneously.

Actual behaviour

Observe HTTP 503 errors and pod restarts because of the OpenProject pod being OOM killed.

Expected behaviour

OpenProject continues to work as expected.

🔄 Requirements for migrations

• Describe manual steps required to update existing deployments. This especially applies if this MR introduces breaking changes:
• Any other considerations in context of the update:

Checklist / Sign-offs

🏷️ Labels

Set labels:

/label ~"MR-Type::Bugfix"
/label ~"PO::👀"
/label ~"QA::👀"
/label ~"Testautomation::👀"

👷 Developer Checklist

Documentation:

Does this MR introduce changes (e.g., new secrets, configuration options) that require documentation?

• No
• Yes, and the documentation has been updated accordingly

Quality Assurance:

• Verified that the feature works as expected, including upgrade scenarios
• Performed regression testing