refactor(executor): split providers and preserve legacy entrypoints

feat: passthrough upstream response headers to clients

refactor(all): replace manual pointer assignments with `new` to enhance code readability and maintainability

refactor(translator, executor): remove redundant `bytes.Clone` calls for improved performance

refactor(executor): optimize payload cloning and streamline SDK translator usage

feat(openai): add responses/compact support

feat(executor): add request_id to error logs

feat(executor): apply payload rules using requested model

feat(registry): support provider-specific model info lookup

refactor(thinking): pass source and target formats to ApplyThinking for cross-format validation

fix(executor): properly handle thinking application errors

refactor: improve thinking logic

feat(executor): add HttpRequest support across executors for better http request handling

feat(executor): enhance payload translation with original request context

fix(thinking): fallback to upstream model for thinking support when alias not in registry

fix(thinking): use model alias for thinking config resolution in mapped models

refactor(executor): resolve upstream model at conductor level before execution

fix(executor): use upstream model for thinking config and payload translation

fix: apply thinkingLevel from model suffix metadata for Gemini 3

Merge pull request #449 from sususu98/fix/gemini-cli-429-retry-delay-parsing

refactor(executor): clarify executor comments and oauth names

refactor(executor): clarify providers and streams

fix(gemini-cli): enhance 429 retry delay parsing

fix(executor): increase buffer size for stream scanners to 50MB across multiple executors

fix(executor): centralize default thinking config

feat(executor): normalize thinking budget across all Gemini executors

Fixed: #414

refactor(executor): dedupe thinking metadata helpers across Gemini executors

**refactor(executor): simplify Gemini CLI execution and remove internal retry logic**

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

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

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

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

fix(executor): remove outdated Gemini model previews from CLI fallback order

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

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

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

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

Refactor executor error handling and usage reporting

Enhance logging for API requests and responses across executors

Feature: #103

feat(registry, executor, util): add support for `gemini-2.5-flash-image-preview` and improve aspect ratio handling

feat(registry, executor): add support for `glm-4.6` model and enhance Gemini CLI token handling

feat(registry, executor): add support for `gemini-2.5-flash-image` model

fix(gemini): Disable thinking config for incompatible models