This is the sub-task for the group impersonation work. https://issues.redhat.com/browse/CONSOLE-4734 

This task is to enhance the impersonation modal by adding an autocomplete feature to both the Username and Group input fields. This will improve usability and reduce the chance of typos. The design and layout should be based on the provided Figma prototype.

Acceptance Criteria:

• When a user types 3 or more characters into the "Username" field, a dropdown list of matching users is displayed.

• When a user types 3 or more characters into the "Group" field, a dropdown list of matching groups is displayed.

• Selecting a name from the list populates the input field.

• The following 2 are still in discussion:
  • The autocomplete query must respect the permissions of the logged-in administrator.

• • If the administrator lacks the necessary permissions to list users or groups, the autocomplete feature should fail gracefully (i.e., no suggestions appear, and no disruptive errors are shown in the UI). 

Some concerns to consider:

1. Permission Check Strategy: How can the frontend reliably determine if the current user has list permissions for users and groups on the cluster? Should we perform a SelfSubjectAccessReview check before enabling the autocomplete feature?

1. API Endpoints: What are the appropriate Kubernetes API endpoints to query for a live search of users and groups?

1. Service Account Authority: Does the OpenShift console's backend service account have the necessary authority to perform these queries on behalf of the user? We must ensure our backend isn't blocked from making the list calls due to its own permission limitations.

1. Performance:

• • Implement debouncing on the input fields to prevent excessive API calls while the user is typing.

• • Ensure the API queries are efficient and filter results server-side where possible to minimize payload size.