plan: automatic gemini provider support

[Copilot]

Planning document for adding Google Gemini as a first-class inference provider with automatic credential discovery.

Upstream check (NVIDIA/OpenShell)

Provider	Status	Action
Copilot	PR #476 by @htekdev — CLI agent provider with credential discovery, network policy, binary detection	Skip
Gemini	No existing work	Proceed
Vertex AI	#472 — credential refresh trait, different auth model	No conflict

Gemini plan

• New gemini provider type using Google's OpenAI-compatible endpoint (generativelanguage.googleapis.com/v1beta/openai/)
• Credential discovery from GEMINI_API_KEY / GOOGLE_API_KEY
• ****** OpenAI-compatible protocols — zero router changes
• ~80 LOC across 5 files (1 new gemini.rs, 4 edits to registry/profiles/CLI)

Open questions

1. API key only for v1, or include OAuth2?
2. Keep Gemini and Vertex AI as separate provider types?

---

📍 Connect Copilot coding agent with Jira, Azure Boards or Linear to delegate work to Copilot in one click without leaving your project management tool.