Add support for Custom Model to OpenAI settings

[cmgriffing]

---

Description

This PR adds a new setting to VSCode in the AI section for a custom model name to be sent to an OpenAI-compatible API. A larger PR was in the works until I realized that the existing custom URL functionality would work.

Closes #3223

Checklist

• I have followed the guidelines in the Contributing document
• My changes follow the coding style of this project
• My changes build without any errors or warnings
• My changes have been formatted and linted
• My changes include any required corresponding changes to the documentation (including CHANGELOG.md and README.md)
• My changes have been rebased and squashed to the minimal number (typically 1) of relevant commits
• My changes have a descriptive commit message with a short title, including a Fixes $XXX - or Closes #XXX - prefix to auto-close the issue that your PR addresses

[cmgriffing]

I have not made any changes to the CHANGELOG or README, yet. I am happy to do so after we have agreed on the approach in this PR.

I have a different version of this PR that adds a CustomProvider class, but after some thought, it might not be necessary since the custom URL for OpenAI is very close to working with any server (not just Azure).

[cmgriffing-at-gitkraken]

@SpaceMerlin33 Do you mind taking a look at this?

[cmgriffing-at-gitkraken]

@trevor-polidore Let me know what you think.

[eamodio]

@cmgriffing There have been significant changes to the AI providers, can you adapt this to the new model?

[eamodio]

Do we need another setting for the custom model? Can't we just use the model setting above?