chore: add assign-to-agent docs & update ai triage

[mnkiefer]

Workflow Updates

• Updated AI Triage Campaign workflow to filter out issues that already have assignees, ensuring only unassigned issues are processed and routed to the project board.
• For unassigned issues with AI-Readiness Score ≥ 9, the workflow outputs an assign_to_agent safe-output item to immediately assign the issue to @copilot.
• Workflow permissions have been updated to require write access for actions, contents, issues, and pull-requests to support agent assignment.
• Update docs in safe-outputs section with assign-to-agent, detailing its usage, required workflow permissions, supported agents, token requirements, and cross-repository assignment procedures.

Documentation Additions

• Adds a detailed section for the new assign-to-agent safe output, including usage examples and token requirements.

Token Handling Improvements

• The agent assignment workflow in assign_to_agent.go now prioritizes the GH_AW_AGENT_TOKEN secret, which requires specific write permissions, ensuring the correct token is used for agent assignment operations.
• The agent mapping and assignment logic in assign_to_agent.cjs is simplified to only support the Copilot agent, removing references to other agents for now.

[Copilot]

Pull request overview

This PR adds comprehensive documentation for the new assign-to-agent safe output feature and updates the AI triage campaign workflow to use it. The feature enables GitHub Agentic Workflows to automatically assign GitHub Copilot agents to issues that meet AI-readiness criteria.

Key changes:

• Adds complete documentation for assign-to-agent safe output including permission requirements, token configuration, and cross-repository support
• Updates AI triage workflow to filter unassigned issues and automatically assign high-readiness issues (score ≥ 9) to @copilot
• Adds new assign_to_agent job to compiled workflow with proper GraphQL mutations and fallback handling

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 3 comments.

File	Description
docs/src/content/docs/reference/safe-outputs.md	Adds comprehensive documentation for the assign-to-agent safe output, including permission requirements, token configuration options, agent output format, supported agents, repository settings, and cross-repository usage
.github/workflows/ai-triage-campaign.md	Updates workflow to filter unassigned issues and automatically assign high-readiness issues to @copilot, adds assign-to-agent configuration, and updates permissions to support agent assignment
.github/workflows/ai-triage-campaign.lock.yml	Compiled workflow file with new assign_to_agent job implementing GraphQL-based agent assignment with permission error handling and fallback mutations

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.