Support Native Kimi K2 Thinking (#1663)

* [feat]: fix k2 prefill

* Update Kimi-K2-Thinking.md

* Create Kimi-K2-Thinking-Native.md

* Update Kimi-K2-Thinking.md

* Update Kimi-K2-Thinking.md

* Update Kimi-K2-Thinking-Native.md

* [perf] optimize K2 MoE weight loading with per-expert pointers

- Avoid expensive torch.stack().contiguous() in Python (was ~6.6s)
- Use per-expert pointer arrays (gate_projs) instead of contiguous memory
- C++ worker pool performs parallel memcpy for TP slicing
- Add LOAD_TIME_PROFILE for load_weights timing analysis

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <[email protected]>

---------

Co-authored-by: ouqingliang <[email protected]>
Co-authored-by: Claude <[email protected]>

Author: 3 people

doc/en/Kimi-K2-Thinking-Native.md

@@ -0,0 +1 @@
+需要先写如何安装运行，然后写一个性能，然后链接到如何使用 claude code 接入的文档。

doc/en/Kimi-K2-Thinking.md

@@ -1,4 +1,5 @@
 # KTransformers+SGLang Inference Deployment
+Please Note This is Quantization Deployment. For Native Kimi K2 Thinking deployment please refer to [here](./Kimi-K2-Thinking-Native.md).
 
 ## Installation