Add video input file support

[arnodirlam]

What this does

Enables RubyLLM to recognize and handle video files as attachments, in addition to existing support for images, audio, PDFs, and text.

Why?

• Modern LLMs and AI applications increasingly support video input, making it important for RubyLLM to handle video files natively.
• This enhancement improves the flexibility and completeness of RubyLLM for users who need to process or analyze video content.

Type of change

• Bug fix
• New feature
• Breaking change
• Documentation
• Performance improvement

Scope check

• I read the Contributing Guide
• This aligns with RubyLLM's focus on LLM communication
• This isn't application-specific logic that belongs in user code
• This benefits most users, not just my specific use case

Quality check

• I ran overcommit --install and all hooks pass
• I tested my changes thoroughly
• I updated documentation if needed
• I didn't modify auto-generated files manually (models.json, aliases.json)

API changes

• Breaking change
• New public methods/classes
• Changed method signatures
• No API changes

Related issues

Fixes #259

[arnodirlam]

I've only tested this with Gemini 1.5 Flash and it works.

Updating the models and running the tests makes a lot of errors fail.

What's the best way to proceed from here?

[crmne]

Hey @arnodirlam I'd love to add this to RubyLLM. Do you have the means to test things out with API keys? There's one little thing to change too.

[crmne]

Should rather be in chat_content_spec.rb.

[arnodirlam]

Done in 228ab17.

[crmne]

Oh and check off the items that you did in the quality check section of the PR. All are required.

[arnodirlam]

I've addressed the remaining points now.

One thing left: vision is ambiguous now, because it sometimes implies image and video, but in reality only implies image. In fact, only Google Gemini models have video support.

Should I add methods supports_image? and supports_video? and soft-deprecate supports_vision??

EDIT: Just realized vision implies image and pdf in all cases, so vision is totally appropriate.

[arnodirlam]

Added a supports_video? helper ad clarified some comments and docs. This should be good to merge now 👍