Coverage calculation for this build was skipped because it contains the same data as another build. See this build for the coverage details.
If you feel that something is wrong here, please contact support.

feat(custom): add endpoint-type configuration for custom providers

add endpoint-type option to allow custom OpenAI-compatible providers
to specify which API endpoint to use (auto, responses, chat_completions).
this enables GPT-5 support via custom providers using the v1/responses endpoint.

changes:
- add EndpointType field to customOpenAIProvider and CustomSpec structs
- support endpoint-type in CLI flags, environment variables, and legacy config
- add validation with go-flags choice constraint
- refactor parseCustomProvidersFromEnv to reduce cyclomatic complexity
- add comprehensive tests for all configuration methods
- update README with examples showing endpoint-type usage

default: chat_completions (maintains backward compatibility)