Support GitHub Copilot CLI as a Tier 3 evaluation agent

[lucacavallaro]

SkillEvaluator currently documents support for claude-code, codex, and opencode. Adding Copilot would allow users to measure how agent skills affect Copilot’s behavior using the same baseline comparisons, datasets, quality gates, and reports available for the existing agents.

Possible usage could look like:

skillevaluator doctor --agents copilot --env-mode docker
skillevaluator tier3 evaluate ./my-skill --agents copilot

Is support for GitHub Copilot already planned? If not, would the maintainers accept a contribution implementing a Harbor adapter and registering copilot as an agent?