Freeway: self-hosted OpenAI-compatible gateway for free-tier providers

One endpoint in front of every free LLM tier, with automatic failover and no
provider key ever leaving the box.

Providers are data, not code: adding one is a JSON file plus an env var, never
a TypeScript change. Ships 12 providers pre-written; `verify-provider --apply`
derives the live model list from your own key.

- core: registry, bucketed quota meter (rolling + UTC-calendar windows),
  key-pool store, 6-step model router with diagnostic rejection reasons
- gateway: node:http, failover that commits to a stream only after 200 headers,
  SSE passthrough with usage sniffing, virtual keys with admin/inference roles
- context engine: sqlite sessions and a 5-rung refit ladder, so a 128k thread
  keeps working when it fails over to an 8k model
- cache: exact (sha256, safe-by-default) + opt-in semantic (flat Float32Array)
- guard: inbound limits, quota forecasting, limit calibration from upstream
  headers, key health, secret/PII/injection scanning
- dashboard: one self-contained HTML file, live SSE request log, token usage
- sdk: zero-dep typed client, drop-in for the OpenAI SDK

Zero runtime dependencies. 326 tests, 27-check smoke suite against a scripted
mock upstream so routing is testable without spending quota.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

providers/github-models.json

@@ -0,0 +1,17 @@
+{
+  "$schema": "./_schema.json",
+  "id": "github-models",
+  "label": "GitHub Models",
+  "docs": "https://docs.github.com/en/github-models",
+  "console": "https://github.com/settings/tokens",
+  "priority": 45,
+  "baseUrl": "https://models.github.ai/inference",
+  "auth": { "type": "bearer", "envKeys": ["GITHUB_MODELS_TOKEN", "GITHUB_TOKEN"] },
+  "limits": { "rpm": null, "rpd": null },
+  "limitsSource": "unverified",
+  "verifiedOn": "2026-08-16",
+  "modelsEndpoint": "/models",
+  "modelsPath": "[].id",
+  "notes": "Free for anyone with a GitHub account; needs a PAT with the models:read scope. Limits are tiered by account and not published as flat numbers, hence nulls — calibration learns them. Note modelsPath is [].id: the catalog is a bare array, not an OpenAI {data:[...]} envelope.",
+  "models": []
+}