Register the worker with PipelineRunner.add_workers() before calling
run() instead. The worker argument still works but now emits a
DeprecationWarning and will be removed in a future release.
Update the runner docstrings, the run_test() helper, and all examples
(including the asyncio.gather() forms) to use the new pattern.
Adds InceptionLLMService, an OpenAI-compatible service for Inception's
Mercury-2 diffusion-based reasoning model. Supports reasoning_effort
(instant/low/medium/high) and realtime mode for reduced TTFT.