feat: ix_full_bridge.so — dlopen bridge for ixformer::infer C++ API

Ported from ex_engine/csrc/ix_full_bridge_v2.cpp + ix_moe_bridge.cpp.
Source: upstream_ref/xllm_latest/core/kernels/ilu/ixformer.h

Exposes 14 ixformer::infer functions as Python-callable torch extension:
  Attention: paged_attention, flash_attn_prefill, reshape_and_cache
  MoE: topk_softmax, moe_gen_idx, moe_expand_input, group_gemm,
       moe_combine_result, fused_moe_forward
  Activation: silu_and_mul
  Norm: rms_norm, fused_add_rms_norm
  Linear: linear
  RoPE: rotary_embedding

Build: torch.utils.cpp_extension.load() in docker build (patch_ops.sh)
Links against libixformer.so from base image at runtime.

This replaces PyTorch MoE fallback (the #1 performance bottleneck).
Without bridge: MoE loops over experts in Python → ~3 TPS decode
With bridge: fused 7-step pipeline in C++ → ~16 TPS decode (sub168 level)

qwen3_6_scripts/patch_ops.sh

@@ -260,6 +260,10 @@ else
     echo "[WARN] corex clang++ not found — skipping extension builds"
 fi
 
+build_stage "compiling ixformer bridge .so (MoE + Attention + Norm)"
+bash ./build_ix_bridge.sh "${VLLM_ROOT}" || \
+    echo "[WARN] ix_full_bridge build failed — MoE will use PyTorch fallback"
+
 build_stage "compiling submission Python sources"
 find . -path './wheels' -prune -o -name '*.py' -print0 | xargs -0 python3 -m py_compile
 build_stage "patch script completed"