[Feature] Reorganize all the docs

[zhaochenyang20]

Checklist

• [x] 1. If the issue you raised is not a feature but a question, please raise a discussion at https://github.com/sgl-project/sglang/discussions/new/choose Otherwise, it will be closed.
• [x] 2. Please use English, otherwise it will be closed.

Motivation

1. Quick Start: Sending Requests move to Getting Started

3. print_highlight

if is_in_ci, use html. else directly print it.

4. differentiate two streaming https://docs.sglang.ai/backend/send_request.html

5. add description to this docs:

6. https://docs.sglang.ai/backend/openai_api_completions.html add link to it

7. change meta-llama/Meta-Llama-3.1-8B-Instruct to meta-llama/Llama-3.2-1B-Instruct

8. deepseek

This should link to deepseek official and saying like:

SGLang is recognized as one of the top engines for [DeepSeek model inference](deepseek official). Refer to[ installation and launch](https://github.com/sgl-project/sglang/tree/main/benchmark/deepseek_v3#installation--launch) to fire up DeepSeek V3/R1 on SGLang.

10. https://docs.sglang.ai/backend/openai_api_vision.html

Add link to this https://github.com/sgl-project/sglang/blob/main/examples/runtime/engine/offline_batch_inference_vlm.py to tell users we can use engine to do generate.

11. https://docs.sglang.ai/backend/offline_engine_api.html move this to examples and give links at the beginning.

12. https://docs.sglang.ai/backend/function_calling.html remove this or give link to what is function calling

13. https://docs.sglang.ai/frontend/frontend.html

We can review simon's PR.

14. https://docs.sglang.ai/router/router.html

remove the $

15. https://docs.sglang.ai/references/hyperparameter_tuning.html

Use router if you are not using the engine.

Tell them how to use the least VRAM.

Link to server args.

16. https://docs.sglang.ai/references/accuracy_evaluation.html

Let's do it next time. Link to benchmark and add readme in benchmark.

17. https://docs.sglang.ai/references/multi_node.html

18. https://docs.sglang.ai/references/quantization.html

19. references: move some part to backend.

20. take AMD back:

https://docs.sglang.ai/references/amd_configure.html

Related resources

No response