feat: claude-code backends + llamaswap provider + dogfood the CC engine (#3)

Phase 2: bump majordomo to latest and wire its new llamaswap provider
into gadfly's endpoint switches; add claude-code/sonnet to gadfly's own
dogfood swarm (pin :sha-86f12c1, map CLAUDE_CODE_OAUTH_TOKEN) so the
Phase-1 engine runs as a live competitor; document the Ollama-through-CC
ANTHROPIC_BASE_URL proxy path as example-only.

The 11-model swarm (incl. claude-code/sonnet) reviewed it; 52 findings
graded via the MCP. Folded in the two real ones: a llamaswap
endpointProvider test (caught by claude-code/sonnet, citing CLAUDE.md)
and adding "openai-compatible" to the provider error messages (gpt-oss).

gofmt clean, go vet quiet, go build + go test -race green.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Steve Dudenhoeffer <steve@stevedudenhoeffer.com>
Co-committed-by: Steve Dudenhoeffer <steve@stevedudenhoeffer.com>

examples/claude-code.yml

@@ -60,6 +60,13 @@ jobs:
           # GADFLY_CLAUDE_PERMISSION_MODE: plan          # read-only; never edits
           # GADFLY_CLAUDE_ALLOWED_TOOLS: "Read,Grep,Glob"
           # GADFLY_CLAUDE_EXTRA_ARGS: "--max-turns 30"
+          # Alternate backend (EXAMPLE ONLY, not validated): point Claude Code at
+          # an Anthropic-API-compatible proxy (e.g. claude-code-router / LiteLLM in
+          # front of Ollama) to run Ollama models THROUGH the CC harness. The
+          # subprocess env forwards ANTHROPIC_*, so just set these instead of the
+          # token above. Tool-use support depends on the proxy/backend.
+          # ANTHROPIC_BASE_URL: ${{ vars.ANTHROPIC_BASE_URL }}
+          # ANTHROPIC_AUTH_TOKEN: ${{ secrets.ANTHROPIC_AUTH_TOKEN }}
           GADFLY_ALLOWED_USERS: "your-username"
           # --- event context (leave as-is) ---
           EVENT_NAME: ${{ github.event_name }}