Feat/image input compression

[afjcjsbx]

📝 Description

This PR adds configurable inbound image compression for PicoClaw's vision pipeline.

Previously, inbound images from channels were only constrained by max_media_size, with no configurable multi-level compression policy before building the model payload. This could lead to oversized inline image payloads and unnecessary pressure on multimodal providers.

With this change, PicoClaw now supports a new agents.defaults.image_input configuration block that allows:

• enabling or disabling automatic inline attachment of user images
• choosing a compression preset (off, low, balanced, aggressive, extreme)
• bounding inline payload size
• resizing images with max width / height limits
• tuning JPEG quality
• selecting the target output format (auto, jpeg, png)

The implementation also preserves the existing local path-tag behavior, so images remain accessible through file references while vision-capable providers can receive a compressed inline image payload when enabled.

🗣️ Type of Change

• 🐞 Bug fix (non-breaking change which fixes an issue)
• ✨ New feature (non-breaking change which adds functionality)
• 📖 Documentation update
• ⚡ Code refactoring (no functional changes, no api changes)

🤖 AI Code Generation

• 🤖 Fully AI-generated (100% AI, 0% Human)
• 🛠️ Mostly AI-generated (AI draft, Human verified/modified)
• 👨‍💻 Mostly Human-written (Human lead, AI assisted or none)

🔗 Related Issue

N/A

📚 Technical Context (Skip for Docs)

• Reference URL: N/A
• Reasoning: The previous inbound media flow did not provide a configurable compression strategy for user-supplied images before sending them to multimodal models. This PR introduces a production-ready, config-driven image optimization layer that reduces the risk of oversized payloads while preserving compatibility with the existing media resolution flow.

🧪 Test Environment

• Hardware:
• OS:
• Model/Provider:
• Channels:

📸 Evidence (Optional)

Click to view Logs/Screenshots

☑️ Checklist

• My code/docs follow the style of this project.
• I have performed a self-review of my own changes.
• I have updated the documentation accordingly.