sycl: add FILL, CUMSUM, DIAG, SOLVE_TRI, SSM_SCAN, GATED_DELTA_NET

[aicss-genai]

Overview

Authors

• @unrahul
• @ctao456
• @malsbat
• @YuningQiu
• @vvravi1
• @zhiqitao-intel

Adds SYCL implementations for six ops previously missing from the backend:

• GGML_OP_FILL
• GGML_OP_CUMSUM
• GGML_OP_DIAG
• GGML_OP_SOLVE_TRI (constrained to N ≤ SYCL_SOLVE_TRI_MAX_N,
  K ≤ SYCL_SOLVE_TRI_MAX_K)
• GGML_OP_SSM_SCAN (two shape regimes: head_dim 128/256 with
  ne1 % WARP_SIZE == 0, and the mamba-1 head_dim=16 case)
• GATED_DELTA_NET (wired via new header)

Each op lives in its own .cpp/.hpp pair under ggml/src/ggml-sycl/.
Dispatch added to ggml_sycl_compute_forward and the shape/type gating added to ggml_backend_sycl_device_supports_op.

Additional information

Part of work originally submitted as #22066 — splitting per reviewer request.

Note: GGML_OP_GATED_DELTA_NET may overlap with parallel upstream work (#20455). Happy to rebase or drop this op from the PR if a canonical implementation is preferred.

Requirements

• I have read and agree with the contributing guidelines
• AI usage disclosure: Yes. This work was partially produced with an agentic engineering approach: agents surface issues and explore experiments while engineers identify and reject candidates using domain knowledge. Human feedback involved.

[ggml-gh-bot]

Hi @aicss-genai, thanks for your contribution!

Per our contribution guidelines, the automated PR checker found the following issue(s) that need your attention:

• Multiple open PRs from a new contributor: We limit new contributors (those without a previously merged PR) to 1 open PR at a time. You currently have 5 open PRs.

• AI-generated content: This project does not accept PRs, descriptions or commit messages that are fully or predominantly AI-generated. If you have used AI to assist you in writing code, please make sure to disclose that explicitly.

---

Please note that maintainers reserve the right to make final decisions on PRs. If you believe there is a mistake, please comment below.