test: run examples end to end against the live API

[jbeckwith-oai]

Summary

• add a workflow_dispatch-only suite that runs 22 Ruby examples against the live OpenAI API
• collect and upload SimpleCov HTML/JSON reports, with a checked-in 89.96% Ruby 4.0 line-coverage floor
• classify every example as covered or explicitly excluded, and validate expected output
• update image examples and README usage to gpt-image-2

Fixes uncovered by the live suite

• accept null finish_reason values in intermediate legacy completion stream chunks, including RBI/RBS updates and a regression test
• use response_id when resuming response streams

Impact

The live suite is opt-in and only runs through manual workflow dispatch. A dispatched run fails if an example fails, the inventory becomes incomplete, or SDK line coverage falls below the current baseline.

Validation

• 22/22 runnable examples passed against the live API on Ruby 4.0
• SimpleCov: 89.96% (28,000/31,122 lines)
• 629 tests, 2,356 assertions
• 2,617 files passed RuboCop
• Sorbet and RBS validation passed
• Ruby 4.0 coverage-regression test verifies a below-baseline run exits non-zero while retaining its JSON report

[openai-sdks]

OkTest Summary

✅ 237/237 SDK tests passed in 12.08s for Ruby SDK PR #374.

Test results — 42 files

Test	Result	Time
tests/chat-completions-complex-body.test.ts	✅ Passed	399ms
tests/chat-completions-create.test.ts	✅ Passed	426ms
tests/chat-completions-stream.test.ts	✅ Passed	208ms
tests/files-content-binary.test.ts	✅ Passed	357ms
tests/files-create-multipart.test.ts	✅ Passed	471ms
tests/files-list-pagination.test.ts	✅ Passed	226ms
tests/initialize-config.test.ts	✅ Passed	428ms
tests/instance-isolation.test.ts	✅ Passed	321ms
tests/models-list.test.ts	✅ Passed	414ms
tests/responses-background-lifecycle.test.ts	✅ Passed	339ms
tests/responses-body-method-errors.test.ts	✅ Passed	603ms
tests/responses-cancel-timeout.test.ts	✅ Passed	279ms
tests/responses-cancel.test.ts	✅ Passed	363ms
tests/responses-compact-retries.test.ts	✅ Passed	441ms
tests/responses-compact.test.ts	✅ Passed	388ms
tests/responses-create-advanced-stream.test.ts	✅ Passed	208ms
tests/responses-create-advanced.test.ts	✅ Passed	295ms
tests/responses-create-disconnect.test.ts	✅ Passed	416ms
tests/responses-create-errors.test.ts	✅ Passed	431ms
tests/responses-create-malformed-api-responses.test.ts	✅ Passed	372ms
tests/responses-create-retries.test.ts	✅ Passed	550ms
tests/responses-create-stream-failures.test.ts	✅ Passed	225ms
tests/responses-create-stream-timeout.test.ts	✅ Passed	233ms
tests/responses-create-stream-wire.test.ts	✅ Passed	2.329s
tests/responses-create-stream.test.ts	✅ Passed	159ms
tests/responses-create-terminal-states.test.ts	✅ Passed	384ms
tests/responses-create-timeout.test.ts	✅ Passed	329ms
tests/responses-create.test.ts	✅ Passed	381ms
tests/responses-delete.test.ts	✅ Passed	318ms
tests/responses-input-items-errors.test.ts	✅ Passed	295ms
tests/responses-input-items-list.test.ts	✅ Passed	221ms
tests/responses-input-items-options.test.ts	✅ Passed	278ms
tests/responses-input-tokens-count-timeout.test.ts	✅ Passed	396ms
tests/responses-input-tokens-count.test.ts	✅ Passed	400ms
tests/responses-malformed-inputs.test.ts	✅ Passed	2.563s
tests/responses-not-found-errors.test.ts	✅ Passed	486ms
tests/responses-parse.test.ts	✅ Passed	239ms
tests/responses-retrieve-retries.test.ts	✅ Passed	394ms
tests/responses-retrieve.test.ts	✅ Passed	355ms
tests/responses-stored-method-errors.test.ts	✅ Passed	1.177s
tests/retry-behavior.test.ts	✅ Passed	2.12s
tests/sdk-error-shape.test.ts	✅ Passed	569ms

View OkTest run #31635745495

_{SDK merge (bc85b79a583a) · head (2234eda431f7) · base (919977919667) · OkTest (a845fa206fa4)}