Feature control for Ai Agents

When checking access for ai agents, we only check in https://gitlab.com/gitlab-org/gitlab/-/blob/master/ee/app/policies/ee/project_policy.rb#L214-214 whether the feature is enabled and the license is correct, but not weather the user has the correct access to the project. We also don't allow setting the project settings of ai agents to private/disabled/public.

To fix this, we need to add feature control for ai agents, similar to how we did for model registry: #412734 (closed)

Steps:

• Add configuration to project and database - Adds backend project setting for model registry (!138399 - merged)
  • Example: Adds model for model_experiments_access_level (!121650 - merged)
• Update read and write policies permissions policies - Adds project feature availability check for mod... (!138995 - merged)
  • Examples:
  • Creates policy for model experiments access (!121654 - merged)
  • Adds write policy for Model experiments (!126041 - merged)
• Add API endpoint to control project setting - Adds API to control model registry project setting (!139004 - merged)
  • Example: Resolve "Add API for enabling/disabling model e... (!132972 - merged)
• Add frontend UI
  • Example: Adds frontend to toggle visibility of model exp... (!121653 - merged)