Using Chat Completions

[alicecarra]

Resolve error described in #16

[alicecarra]

Needs better null handling

[Copilot]

Pull request overview

This PR updates the OpenAI integration to use the Chat Completions endpoint, addressing Issue #16 where chat models were being sent to the legacy /v1/completions endpoint and returning a 404.

Changes:

• Switch OpenAI requests from completions.create to chat.completions.create and update response parsing accordingly.
• Bump the extension version and upgrade the openai npm dependency.
• Update package-lock.json to reflect dependency graph changes from the openai upgrade.

Reviewed changes

Copilot reviewed 2 out of 3 changed files in this pull request and generated 2 comments.

File	Description
src/openai-service.ts	Migrates revision/translation calls to chat.completions and reads output from choices[0].message.content.
package.json	Bumps extension version to 0.0.7 and upgrades openai to ^4.70.0.
package-lock.json	Updates locked openai version and transitive dependency set after upgrade.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

response.choices[0].message.content can be null/undefined (and choices[0] can be missing). Returning the literal fallback string "null response lol" will insert user-visible garbage and is unprofessional. Please add a proper guard (e.g., check choices.length and message.content) and either return undefined per the interface or throw a descriptive error/message when the API returns no content.

[Copilot]

Leaving a TODO-style comment (// FIX ME: Not the best null handling...) alongside the placeholder fallback string makes this behavior knowingly incorrect. Please implement proper empty-response handling here as well (guard missing choices/content and return undefined or surface a clear error to the user), and remove the TODO/placeholder text.