add support for Ollama num_gpu

[Fmstrat]

Description

Added support for Ollama num_gpu. This allows for forcing models to use the GPU in memory limited situations. Modeled after existing Ollama specific options such as keepAlive.

Checklist

• The relevant docs, if any, have been updated or created
• The relevant tests, if any, have been updated or created

Testing instructions

Set your config up with numGpu:

    {
      "title": "Qwen2.5 Coder 14B (Ollama)",
      "provider": "ollama",
      "model": "qwen2.5-coder:14b",
      "apiBase": "http://localhost:11434",
      "contextLength": 2048,
      "completionOptions": {
          "maxTokens": 1024,
          "numGpu": 1000
      },
      "keepAlive": 0
    },

Watch is pass through to Ollama via the Ollama logs (--n-gpu-layers 1000):

level=INFO source=server.go:376 msg="starting llama server" cmd="/usr/lib/ollama/runners/cuda_v12_avx/ollama_llama_server runner --model /root/.ollama/models/blobs/sha256-ac9bc7a69dab38da1c790838955f1293420b55ab555ef6b4615efa1c1507b1ed --ctx-size 2048 --batch-size 512 --n-gpu-layers 1000 --threads 8 --no-mmap --parallel 1 --tensor-split 35,13 --port 33783"

[netlify]

✅ Deploy Preview for continuedev ready!

Name	Link
🔨 Latest commit	abfd726
🔍 Latest deploy log	https://app.netlify.com/sites/continuedev/deploys/68112810d5363400088fe068
😎 Deploy Preview	https://deploy-preview-4353--continuedev.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify site configuration.

[Fmstrat]

I think this is failing due to an unrelated timeout?

[warcow105]

This is a welcome addition. It will allow the use of larger models or deeper context without forcing those changes on all of the models.

[sestinj]

Hi @Fmstrat, yesterday we shared some updates with our contributors about how we're aiming to improve the contribution process. Part of this included the addition of a Contributor License Agreement (CLA) to protect both contributors and the project. We're reaching out to ask that previous contributors sign it.

Could you please take a moment to sign, or if you have any questions send me a message? (either here or nate@continue.dev would work)

To do so, you just need to post a comment below with the following text:

I have read the CLA Document and I hereby sign the CLA

❤️ Thank you for the work you've done on Continue, and let me know if you have any suggestions on how we can make the project even better!

[github-actions]

Thank you for your submission, we really appreciate it. Like many open-source projects, we ask that you all sign our Contributor License Agreement before we can accept your contribution. You can sign the CLA by just posting a Pull Request Comment same as the below format.

---

I have read the CLA Document and I hereby sign the CLA

---

1 out of 2 committers have signed the CLA.
✅ (sestinj)[https://github.com/sestinj]
❌ @Fmstrat
_{You can retrigger this bot by commenting recheck in this Pull Request. Posted by the CLA Assistant Lite bot.}