fix: prepend text message to content blocks in multimodal agent loop

[LupoGrigi0]

Summary

Fixes #1043 — When image attachments are present, the agent loop drops the user's text message. The LLM receives images without any context about what the user asked.

Changes

File: crates/openfang-runtime/src/agent_loop.rs (both streaming and non-streaming paths)

The fix prepends the text message as a ContentBlock::Text into the image blocks vector, so the LLM receives both text and images in a single multimodal turn.

Before (broken)

if let Some(blocks) = user_content_blocks {
    // blocks = images ONLY — text message silently dropped
    session.messages.push(Message::user_with_blocks(blocks));
} else {
    session.messages.push(Message::user(user_message));
}

After (fixed)

if let Some(mut blocks) = user_content_blocks {
    if !user_message.is_empty() {
        blocks.insert(0, ContentBlock::Text {
            text: user_message.to_string(),
            provider_metadata: None,
        });
    }
    session.messages.push(Message::user_with_blocks(blocks));
} else {
    session.messages.push(Message::user(user_message));
}

Testing

Test	Before	After
100x100 blue square + "What color?"	"I can't see the image"	"Blue"
388KB screenshot + "Describe this"	Hallucinated response	Accurate description
1.3MB bird illustration	81K tokens consumed, hallucinated inability	"Stippled illustration of a bird"

• Tested with Qwen 3.5 Plus and Gemini 2.5 Flash via OpenRouter
• Images up to 1.3MB (1.8MB base64) confirmed working
• Direct OpenRouter API calls verified that both models support vision — the issue was purely in the agent loop's message construction
• Fix applied to both non-streaming (run_agent_loop) and streaming (run_agent_loop_streaming) paths
• Running in production across 3 OpenFang instances for the HACS coordination system

---

Submitted by Cairn-2001 ([email protected]), OpenFang maintainer for HACS at smoothcurves.nexus

[jaberjaber23]

Clean, targeted fix for #1043. Inserting the text block at index 0 with the !user_message.is_empty() guard is the right call (avoids a stray empty Text block when the channel bridge sends images without caption).

Same rebase-needed note: CI isn't registered on this branch. Rebase on latest main to trigger checks and we'll merge once green.

[jaberjaber23]

Clean fix. The prepend-text-to-blocks logic is correct. Cannot merge as-is due to conflicts with main (agent_loop.rs has drifted). Please rebase and we'll land it.

[jaberjaber23]

Fix landed in main via the new build_user_turn_message helper in crates/openfang-runtime/src/agent_loop.rs:270-287, called from both dispatch sites (lines 405 and 1633). That helper is a strict superset of this PR: it prepends text at index 0 (same as here), additionally guards with trim().is_empty() so whitespace-only captions are dropped, and handles the empty-blocks-vec case. Regression tests at lines 3606, 3616, 3630, 3649, 3661 cover the cases. Thanks for spotting the original issue.