[devin-ai-integration]

Summary

Closes a code-scan coverage gap: BaseMessage.add_content/add_text/add_image and Discussion.add_image had no tests, so the string→list content promotion and the base64 image URL construction that feed vision-agent prompts could regress silently (malformed or dropped screenshots surface as a quality drop, not an error). Only merge() was tested.

Tests added in tests/llm/test_llm_utils.py, in the style of the existing merge tests:

• add_text on a string-bodied message promotes it to [{type: text, ...}, {type: text, ...}]; add_content appends to an already-list body.
• add_image with a str passes the URL through (and attaches detail when given).
• add_image with a PIL image and with an ndarray (parametrized) yields data:image/jpeg;base64,...; the payload is decoded and reopened to assert it is a real JPEG of the original size.
• Discussion.add_image mutates only the last message.

Devin-Org: engineering

Link to Devin session: https://app.devin.ai/sessions/bf387a97923c43d5843ab8af74cfa268
Open in Devin Desktop: https://app.devin.ai/desktop/session/bf387a97923c43d5843ab8af74cfa268?variant=devin

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment, CI, and merge conflict monitoring