ref(upstream): FULL TREE — Deep-Spark xllm (1470) + ds_vllm csrc/models (703)

Replaces cherry-picked upstream_ref with complete source trees.

xllm/ — Iluvatar official C++ inference engine (15MB, 1470 files)
  Complete: kernels → layers → models → runtime → scheduler → api
  Excluded: .git, binary images, third_party submodule checkouts

ds_vllm/ — Iluvatar official vllm fork (8MB, 703 files)
  Included: csrc/ (ALL CUDA kernels), fused_moe/, qwen3_5 model, _custom_ops
  Excluded: tests, benchmarks, docs, examples (not needed for reference)

Critical call chains now fully traceable:
  MoE: moe_topk_softmax_kernels.cuh → ixformer.h → fused_moe.cpp → layer
  GDN: qwen3_gated_delta_net_base.cpp → qwen3_5_gated_delta_net.cpp
  Attention: ixformer.h → xllm_paged_attention → attention.cpp

upstream_ref/xllm/examples/generate_embedding.py

@@ -0,0 +1,30 @@
+# python examples/generate_embedding.py --model='/path/models/Qwen3-8B' --devices='npu:0' --runner pooling
+# python generate_embedding.py --model='/path/models/Qwen3-8B' --devices='npu:0,npu:1'
+
+from xllm import ArgumentParser, LLM, PoolingParams
+
+# Create an embedding LLM.
+parser = ArgumentParser()
+args = parser.parse_args()
+llm = LLM(**vars(args))
+
+# Create pooling params.
+pooling_params = PoolingParams()
+
+inputs = [
+    "Hello, my name is",
+    "The president of the United States is",
+    "The capital of France is",
+    "The future of AI is",
+]
+
+outputs = llm.embed(inputs, pooling_params=pooling_params)
+
+# Print the outputs.
+for i, output in enumerate(outputs):
+    input_str = output.prompt
+    generated_embedding = output.outputs.embedding
+    print(f"Input: {input_str!r}, Generated embedding: {generated_embedding!r}")
+
+llm.finish()
+