ci : enable hf-jobs on self-hosted server-cuda

[CISC]

Overview

HF jobs is ready, let's start by testing at a small scale to see how it performs.

Requirements

• I have read and agree with the contributing guidelines
• AI usage disclosure: تيدق
• Language disclosure: Malay (Jawi script)

[CISC]

Haha, just found (and fixed) a bug, hf-jobs doesn't accept . in tags, so failed on llama.cpp... :)

[CISC]

Huh, strange, failed, but still running on HF:
https://github.com/ggml-org/llama.cpp/actions/runs/33659873639/job/100347514480

[CISC]

WTF, picking up wrong jobs?

2026-09-02 18:14:02Z: Listening for Jobs
2026-09-02 18:14:09Z: Running job: cpu-x64-high-perf
2026-09-02 18:14:16Z: Job cpu-x64-high-perf completed with result: Failed

Edit: https://github.com/ggml-org/llama.cpp/actions/runs/33665872870/job/100367355041#step:1:2

[CISC]

WTF, picking up wrong jobs?

Restarted server just in case, though this should not be possible unless GitHub is passing out wrong ids, will keep an eye out for any more shenanigans...

[CISC]

Mystery solved: huggingface/jobs-actions#9