Add default_completions_model trait

[srdas]

Description

• Addresses comment in model_parameters config not applied to embeddings model fields in jupyter_jupyter_ai_config.json #1296.
• Fixes Inline completion model provider should be traitlet-configurable #781 (11 upvotes!).
• Follow-up to Pass model_parameters trait to embedding & completion models #1298.

Users can specify a pre-specified configuration file titled jupyter_jupyter_ai_config.json in any of the paths shown when running jupyter --paths (see the documentation here). A sample of this file is shown below:

{
  "AiExtension": {
    "default_language_model": "openrouter:microsoft/phi-4",
    "default_embeddings_model": "openai-custom:BAAI/bge-m3",
    "default_completions_model": "ollama:llama3.2",
    "default_api_keys": {
      "OPENROUTER_API_KEY": "sk-XXXX",
      "OPENAI_API_KEY": "sk-YYYY"
    },
    "model_parameters": {
      "openrouter:microsoft/phi-4": {
        "openai_api_base": "https://chat.example.com"
      },
      "openai-custom:BAAI/bge-m3": {
        "openai_api_base": "https://chat.example.com"
      },
      "ollama:llama3.2": {
        "base_url": "http://localhost:11434"
      }
    }
  }
}

Issue: Currently, the default_completions_model is not being updated in config.json as this is not handled. This PR enables completions models config in a manner consistent with that done for language models and embedding models.

For reviewers

To test this:

1. Create a new file titled jupyter_jupyter_ai_config.json in the path ~/.jupyter/ (or any other path seen from running jupyter --paths). Edit in the content of the file above.
2. Delete the config.json file at path ~/Library/Jupyter/jupyter-ai/
3. Start jupyter lab from the CLI
4. Look at the config.json file to make sure it is also correctly showing the completions_model_provider_id and the completions_fields with the required content from the pre-specified config file.
5. Open the AI setting pane to see all the fields are automatically populated with the content from the file jupyter_jupyter_ai_config.json

[dlqqq]

@srdas Thank you for building this! Changes look good. Verified with local testing. Feel free to merge & backport!

[srdas]

@meeseeksdev please backport to 2.x

[lumberbot-app]

Owee, I'm MrMeeseeks, Look at me.

There seem to be a conflict, please backport manually. Here are approximate instructions:

1. Checkout backport branch and update it.

git checkout 2.x
git pull

2. Cherry pick the first parent branch of the this PR on top of the older branch:

git cherry-pick -x -m1 a82eecc220a032b9bfbef9beb3da900a8bfd4e59

3. You will likely have some merge/cherry-pick conflict here, fix them and commit:

git commit -am 'Backport PR #1303: Add `default_completions_model` trait'

4. Push to a named branch:

git push YOURFORK 2.x:auto-backport-of-pr-1303-on-2.x

5. Create a PR against branch 2.x, I would have named this PR:

"Backport PR #1303 on branch 2.x (Add default_completions_model trait)"

And apply the correct labels and milestones.

Congratulations — you did some good work! Hopefully your backport PR will be tested by the continuous integration and merged soon!

Remember to remove the Still Needs Manual Backport label once the PR gets merged.

If these instructions are inaccurate, feel free to suggest an improvement.