[DEPLOY] Complete submission: baseline + all optimizations

Adds ALL files needed for Dockerfile build:
  - qwen3_6_scripts/ (baseline patches + our optimizations)
  - vllm/ (full vllm package)
  - paged_attention_v2_pytorch.py (V2 with single-bmm optimization)
  - Dockerfile + computility-run.yaml

Our optimizations vs baseline:
  1. paged_attn.py: pre-gathered context KV (eliminates 194 gather calls),
     Triton try/fallback, V2 heuristic, threshold 32K→64K
  2. paged_attention_v2_pytorch.py: fills NotImplementedError,
     single-bmm Phase 1 (195 launches → 3)
  3. patch_enable_triton.py: HAS_TRITON=True with safety fallback
  4. patch_triton_tuning.py: BLOCK=64, NUM_WARPS=4 for BI-V100
  5. computility-run.yaml: gpu-memory-utilization 0.9→0.95,
     max-num-batched-tokens 8192→16384

This repo can now be submitted to dev.modelhub.org.cn as-is.

vllm/transformers_utils/configs/__init__.py

@@ -0,0 +1,40 @@
+from vllm.transformers_utils.configs.chatglm import ChatGLMConfig
+from vllm.transformers_utils.configs.dbrx import DbrxConfig
+from vllm.transformers_utils.configs.eagle import EAGLEConfig
+from vllm.transformers_utils.configs.exaone import ExaoneConfig
+# RWConfig is for the original tiiuae/falcon-40b(-instruct) and
+# tiiuae/falcon-7b(-instruct) models. Newer Falcon models will use the
+# `FalconConfig` class from the official HuggingFace transformers library.
+from vllm.transformers_utils.configs.falcon import RWConfig
+from vllm.transformers_utils.configs.internvl import InternVLChatConfig
+from vllm.transformers_utils.configs.jais import JAISConfig
+from vllm.transformers_utils.configs.medusa import MedusaConfig
+from vllm.transformers_utils.configs.mllama import MllamaConfig
+from vllm.transformers_utils.configs.mlp_speculator import MLPSpeculatorConfig
+from vllm.transformers_utils.configs.mpt import MPTConfig
+from vllm.transformers_utils.configs.nemotron import NemotronConfig
+from vllm.transformers_utils.configs.nvlm_d import NVLM_D_Config
+from vllm.transformers_utils.configs.qwen2vl import (Qwen2VLConfig,
+                                                     Qwen2VLVisionConfig)
+from vllm.transformers_utils.configs.solar import SolarConfig
+from vllm.transformers_utils.configs.ultravox import UltravoxConfig
+
+__all__ = [
+    "ChatGLMConfig",
+    "DbrxConfig",
+    "MPTConfig",
+    "RWConfig",
+    "InternVLChatConfig",
+    "JAISConfig",
+    "MedusaConfig",
+    "EAGLEConfig",
+    "ExaoneConfig",
+    "MllamaConfig",
+    "MLPSpeculatorConfig",
+    "NemotronConfig",
+    "NVLM_D_Config",
+    "SolarConfig",
+    "UltravoxConfig",
+    "Qwen2VLConfig",
+    "Qwen2VLVisionConfig",
+]