fix(executor): prevent streaming on failed response when no fallback

fix(executor): prevent streaming from 429 response when fallback available

feat(executor): add intelligent retry logic for 429 rate limits

**feat(registry): add GPT-5.1 Codex Max model definitions and support**

**feat(auth): add AuthIndex for diagnostics and ensure usage recording**

fix: detect HTML error bodies without text/html content type

feat(auth): add iFlow cookie-based authentication support

**fix(runtime): simplify scanner buffer allocation in executor implementations**

**refactor(runtime): centralize reasoning effort logic for GPT models**

fix(runtime): remove gpt-5.1 minimal effort variant

**feat(runtime): extend executor support for GPT-5.1 Codex and variants**

**feat(runtime): add payload configuration support for executors**

Fixed: #246

feat(gemini-cli): add multi-project support and enhance credential handling

feat(vertex): add usage source resolution for Vertex projects

feat(vertex, management): enhance token handling and OAuth2 integration

feat(vertex): add support for Vertex AI Gemini authentication and execution

fix(executor): ensure usage reporting for upstream responses lacking usage data

feat(config): support HTTP headers across providers

fix(executor): remove default reasoning effort for `gpt-5-codex-mini`

fix(executor): update model name from `codex-mini-latest` to `gpt-5-codex-mini`

add default medium reasoning case for gpt-5-codex-mini

add default medium reasoning case for gpt-5-codex-mini

feat(registry): add GPT-5 Codex Mini model variants

fix(executor): add logging and prompt cache key handling for OpenAI responses

fix(executor): remove outdated Gemini model previews from CLI fallback order

fix(executor): adjust `Anthropic-Beta` header handling for consistent API requests

fix(executor): ensure consistent header application in Claude API requests

fix(executor): remove `safetySettings` from payload in token counting request

fix(executor): remove `safetySettings` from payload for Gemini requests

feat(gemini): add Gemini API key endpoints

refactor(executor): summarize API error bodies of html in debug logs

chore(logging): clarify 429 rate-limit retries in Gemini executor

fix(gemini-cli): correctly strip/normalize thinking config by model

fix(executor): only apply thinking config to supported models

feat(runtime): add User-Agent headers to codex and claude executors

fix(executor): pass authID to relay instead of provider

feat(registry): unify Gemini models and add AI Studio set

Fixed: #172

feat(claude): add model alias mapping and improve key normalization

fix(gemini-executor): uppercase responseModalities

fix(aistudio): ensure colon-spaced JSON in responses

refactor(wsrelay): rename RoundTrip to NonStream

fix(gemini): map responseModalities to uppercase IMAGE/TEXT

fix(aistudio): remove generationConfig and tools when action is countTokens

fix(aistudio): strip usage metadata from non-final stream chunks

feat(aistudio): track Gemini usage and improve stream errors

feat: add websocket routing and executor unregister API

Fixed: #140 #133 #80

feat(executor): add debug logs for rate-limiting retries in Gemini CLI executor