Promote llama.cpp Windows CUDA release in README

[LauraGPT]

Summary

Promotes the published llama.cpp runtime v0.1.7 Windows CUDA asset in the main README surfaces.

• links the exact runtime-llamacpp-v0.1.7 release and funasr-llamacpp-windows-x64-cuda.zip from English and Chinese READMEs
• adds --backend cuda quickstart lines for Windows users
• updates runtime/llama.cpp/README.md from future-tense packaging language to the published v0.1.7 download path
• adds English and Chinese What's New entries for the runtime release

Validation

• doc assertion script confirms all three docs mention runtime-llamacpp-v0.1.7, funasr-llamacpp-windows-x64-cuda.zip, and --backend cuda
• git diff --check

This is documentation-only and follows the successful release workflow at https://github.com/modelscope/FunASR/actions/runs/29657743507.

[gemini-code-assist]

Code Review

This pull request updates the English and Chinese READMEs, as well as the llama.cpp runtime documentation, to announce the release of llama.cpp runtime v0.1.7 and document the new prebuilt Windows CUDA package for SenseVoiceSmall. The review feedback focuses on improving the technical accuracy of the CUDA-related documentation. Specifically, it suggests clarifying that the prebuilt package is compatible with RTX 30-class and newer GPUs due to backward compatibility, and refining the terminology from 'architecture 86' to 'compute capability 8.6' (or 'Ampere architecture').

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.

[gemini-code-assist]

The prebuilt CUDA binary can also run on newer NVIDIA GPUs (such as RTX 40-class / Ada Lovelace) due to backward compatibility. Specifying "RTX 30-class (or newer)" would be more accurate and prevent users with newer GPUs from thinking they cannot use this package.

Suggested change

	# Use the windows-x64-cuda package on RTX 30-class GPUs:
	# Use the windows-x64-cuda package on RTX 30-class (or newer) GPUs:

[gemini-code-assist]

The term "架构 86" (architecture 86) is non-standard in CUDA terminology. It is more accurate to refer to it as "Ampere 架构 (算力 8.6)" or "sm_86 架构". Also, mentioning "or newer" (及以上) is helpful since newer GPUs (like RTX 40-series) are backward-compatible.

Suggested change

	# RTX 30 系列等架构 86 GPU 可使用 windows-x64-cuda 包：
	# RTX 30 系列及以上（Ampere 架构，算力 8.6+）GPU 可使用 windows-x64-cuda 包：

[gemini-code-assist]

In CUDA terminology, "CUDA architecture 86" is more precisely referred to as "CUDA compute capability 8.6" or "sm_86 architecture" (which corresponds to the Ampere architecture). Clarifying this will make the technical requirements clearer to users.

Suggested change

	NVIDIA GPUs that match CUDA architecture 86. Download the CUDA ZIP from
	NVIDIA GPUs with CUDA compute capability 8.6 (Ampere architecture). Download the CUDA ZIP from