Improve SYCL doc

[malsbat]

Overview

The main change is to move the SYCL Docker instructions to the common docker.md file: this should improve accessibility for users navigating from the top-level README.md

Additional information

Beyond some small editorial fixes, this also corrects the documented default of the GGML_SYCL_GRAPH build argument.

Requirements

• I have read and agree with the contributing guidelines
• AI usage disclosure: NO

[kha84]

Guys, can I ask you to add few more things to SYCL docs, pls.

Remove from the project-level README.md the mentioning of Nvidia being supported by SYCL

Currently the top project-level README.md page says SYCL backend can be used for Nvidia:

Backend	Target devices
...	...
SYCL	Intel and Nvidia GPU
...	...

but from the recent news I read:

2026.02
Remove support for Nvidia & AMD GPU, because the oneAPI plugin for Nvidia & AMD GPU is unavailable: download/installation channels are out of work. User can't build up the software for Nvidia & AMD GPU.

This is inconsistecy

Add some short performance comparison table for the reference

It would greatly help to newcomers to understand what to expect from performance standpoint, when SYCL is used as a backend, compared say to Vulcan. I'm not expecting a wide-range comparison, but just few typical models would help a lot, like this:

card	model	backend	pp512	tg128
Arc b580	Qwen3.5-9B-Q4_K_M.gguf	Vulcan	???	???
Arc b580	Qwen3.5-9B-Q4_K_M.gguf	SYCL	???	???

Without this table, people will be running llama.cpp with using SYCL backend blindly, questioning if they're reaching the reference performance or not, wasting their time on troubleshooting issues they might not have. Having reference performance numbers will benefit community.

NB: I do not own the Intel card unfortunately, otherwise I would gladly help you to benchmark it

[NeoZhangJianyu]

Agree to change "Intel and Nvidia GPU" to "Intel GPU" in README.md.

"Add some short performance comparison table for the reference": I don't think it's goo idea.
Compare to other backends will make unnecessary trouble.
But we could create a table to list the performance data of LLM on Intel GPU as reference.
It can be handled by another PR.

Thank you!

[malsbat]

Comments addressed, ready for review again

[arthw]

It's good job!

Thank you!