[Feature] Support for Ascend NPU backend

[botieking98]

Motivation

Support for Ascend NPU backend

Modifications

inspired by #3782
and also support run on multiple NPU devices.

Checklist

• Format your code according to the Code Formatting with Pre-Commit.
• Add unit tests as outlined in the Running Unit Tests.
• Update documentation / docstrings / example tutorials as needed, according to Writing Documentation.
• Provide throughput / latency benchmark results and accuracy evaluation results as needed, according to Benchmark and Profiling and Accuracy Results.
• For reviewers: If you haven't made any contributions to this PR and are only assisting with merging the main branch, please remove yourself as a co-author when merging the PR.
• Please feel free to join our Slack channel at https://slack.sglang.ai to discuss your PR.

Test result

python -m sglang.launch_server --model-path Qwen/Qwen2.5-72B-Instruct --device npu --grammar-backend outlines --attention-backend torch_native --tp 16 --port 40000 --host 0.0.0.0 --trust-remote-code

[22dimensions]

We are very happy to see that someone is also interested in sglang on Ascend.

Thanks for moving this forward, would you mind linking #3781 to make more people know background in details? I'd like to work with you together to support Ascend NPU.