fix: simplify hardRestart — remove helper script, spawn directly with EADDRINUSE retry

feat: SSE update progress + reliable restart on Windows

fix: dark mode tabs, StableText bilingual width, self-update reliability

fix: StableText uses both en+zh references for stable button width

fix: model store dual index, desktop UI fallback, dynamic port support

fix: self-update restart with graceful close + 3s timeout fallback

fix: server restart hanging when active SSE connections exist

feat: display secondary rate limit window (weekly cap) in dashboard

refactor: use toErrorStatus() instead of inline duplicate

fix: use real reset duration from 429 response + auto-fallback to next account (#65)

fix: prevent false window rollover resets in syncRateLimitWindow

fix: review improvements — schema walker, status helper, UI i18n, CI version skip

fix: recursive additionalProperties injection + upstream error status propagation

feat: add Test Connection diagnostic button (#63)

feat: add Structured Outputs support (response_format → text.format)

fix: strict plan filtering in model-aware account routing

fix: let curl auto-negotiate Accept-Encoding for OAuth requests

fix: pass through cached_tokens from Codex API (#58)

feat: model-aware multi-plan account routing (#57)

fix: update model catalog — gpt-5.4/5.3-codex removed, gpt-5.2-codex is new flagship

feat: add force_http11 option for proxies without HTTP/2 support (#56)

fix: version display picks higher of git tag vs package.json

feat: editable proxy_api_key from Dashboard settings panel

fix: version display uses git tag + remove inline-grid layout hack

fix: Electron OAuth login + i18n layout shift + update modal (#53)

Merge pull request #52 from icebear0828/fix/oauth-direct-fallback

fix: use getRootDir() for package.json in self-update detection

fix(auth): add direct-connection fallback for OAuth behind proxy

fix: semver comparison + timer cleanup in self-update

feat: auto-release pipeline + multi-mode update system + commit hash display

fix: preserve model name suffixes (-fast, -high, etc.) in API responses

fix: strip service_tier from request body + add OpenAI-Beta header + update gpt-5.4 efforts

docs: add comment explaining stream:true hardcoding in responses route

feat: /v1/responses passthrough endpoint + missing SSE event types + Docker docs fix (#37, #38)

feat: model name suffix system + service_tier + Docker persistence fix (#35)

feat: proxy assignment management page with bulk operations (#34)

fix: update checker shows correct status + version display + dark mode fixes

feat: proxy pool add form with separate protocol/host/port/auth fields

fix: remove wall-clock timeout from streaming SSE to prevent thinking chain cutoff

fix: proxy pool review fixes — security, validation, robustness

feat: add proxy pool for per-account proxy routing

feat: GitHub Star badge + check-for-updates in dashboard header

fix: move /v1/models/catalog route before :modelId param route

feat: add gpt-5.4 + codex-spark support, matrix model selector, extended reasoning efforts

fix: auto-admit new Codex models from backend API

fix: remove unsupported previous_response_id parameter (#30)

refactor: extract startServer() + path abstraction for Electron reuse

feat: add reasoning/thinking output support for OpenAI and Anthropic routes

fix: Anthropic route thinking blocks validation + context token display

fix: support thinking/redacted_thinking content blocks in Anthropic schema