Feature Requests

> WHAT: Trend-based automatic resource class scaling (opt-in) < The ability for CircleCI to automatically scale a job up to the next resource class based on that job's historical usage trends. If a specific CI job shows memory or CPU usage trending upward over time — i.e., consistently maxing out its current resource class — CircleCI would auto-bump it to the next class up without requiring a manual config change. > WHY: < As usage scales — steadily increasing volume, new services and projects added over time — teams want jobs to right-size themselves automatically, avoiding the manual overhead of monitoring and adjusting resource classes as workloads grow.

When I change a filter, such as "My Pipelines" or the project I'm on, it resets the other filters. This is very disorienting.

At the moment, create a new git-branch in Bitbucket is triggering automatic Circle-CI build although no commit have been pushed to this branch yet. This resulting in a redundant build and in-case your pipeline resides on commit-it it might override resources using this commit-it. CCI-I-684

For our workflow, there are other branches in addition to the default branch that we would like to disable auto-cancel redundant builds for. CCI-I-1584

As GrapheneOS is becoming more popular it would be helpful if an image was provided for testing on it.

We would like to use Project API Token to trigger build as using Personal API Token requires user to login and logout or revoke OAuth when the authentication go stale.

Currently machine executors favour memory increases over CPU. For many jobs more CPU power is required but not a large amount of more memory. It would be useful for CircleCI to offer resource classes that favoured a higher CPU share over high memory for jobs that are not memory intensive but still compute intensive.

It would be incredibly helpful if we could set and manage RBAC like controls for CircleCI user/groups within the platform, similar to how there is enhanced controls and visibility for Argo Rollouts. Keeping management related to CircleCI within CircleCI (or at least visibility to configuration) will improve security and platform management

Gen2 resources are available for Docker and Machine x86 executors, but not yet available for ARM executors

Use Case: We have some Jobs in our workflows, eg linting of code. If these fail the whole workflow shows as Failing, forever. Is there any way we can mark certain Jobs as not important? Would be nice if these non-important jobs would show with status= warning and not failed.Similar (but not the same) as: https://circleci.com/ideas/?idea=CCI-I-216 CCI-I-646