examples/models/qwen3_5_moe: CUDA Engine/Session adapter + OpenAI serving

[mergennachin]

Implement Qwen35MoEEngine / Qwen35MoESession (the model-agnostic
LLMEngine / LLMSession contract) over the exported prefill/decode methods.
serving_capacity() reports a single physical session; the model is
hybrid_recurrent with seek() NotSupported (no prefix reuse). main.cpp is a thin
CLI over the engine/session.

OpenAI serving runs process-isolated and model execution stays in C++: serve.py
is the control plane (FastAPI, chat templating, Qwen XML tool parsing,
validation; no CUDA, no pybind) and spawns qwen3_5_moe_worker
(qwen35_moe_worker.cpp), a C++ worker that constructs the engine and one session
and speaks the same JSONL protocol as the generic text worker. Executing the
AOTI CUDA model inside a live asyncio server process segfaults in the int4
matmul; isolating it in a plain worker process makes serving reliable while
loading weights once. Single-slot: concurrent requests queue. Tool calls use the
Qwen XML <function=...> format (QwenFunctionCallDetector).

Review order: qwen35_moe_engine.{h,cpp} (adapter) and main.cpp; then
qwen35_moe_worker.cpp and serve.py (serving); then tests and docs.

Part of #20001

[mergennachin]

Stack from ghstack (oldest at bottom):

• Qwen3.5-MoE CUDA V2 foundation: one model, many isolated sessions #20117
• -> examples/models/qwen3_5_moe: CUDA Engine/Session adapter + OpenAI serving #20043
• extension/llm/server: document pi integration #19999
• extension/llm/server: worker-based OpenAI-compatible HTTP server #19994
• extension/llm/server: serving foundations (schemas, errors, templating, tools) #19993
• extension/llm/runner: Engine/Session C++ core + token-step primitives #19991

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/executorch/20043

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

❌ 5 New Failures, 1 Pending

As of commit 462416d with merge base eeb0646 ():

NEW FAILURES - The following jobs have failed:

• pull / android / run-emulator (gh)
  The process '/usr/bin/sh' failed with exit code 1
• pull / unittest / linux / linux-job (gh)
  RuntimeError: Command docker exec -t 767e9801d9b145094719d4a90ac656c2a6486fed3c66ab56c8d7d6478f0427ce /exec failed with exit code 1
• pull / unittest / macos / macos-job (gh)
  RuntimeError: Command bash /Users/ec2-user/runner/_work/_temp/exec_script failed with exit code 1
• pull / unittest-editable / linux / linux-job (gh)
  RuntimeError: Command docker exec -t 89a3e559d8cdd2d1db3e47ea7b024a43f4c74c219ad81b8e53121b2402813664 /exec failed with exit code 1
• pull / unittest-editable / macos / macos-job (gh)
  RuntimeError: Command bash /Users/ec2-user/runner/_work/_temp/exec_script failed with exit code 1

This comment was automatically generated by Dr. CI and updates every 15 minutes.