[Feature Request] Recommend System Enhancement: Support Multi-Modal Recommendation

[Yuujee]

Required prerequisites

• I have searched the Issue Tracker and Discussions that this hasn't already been reported. (+1 or comment there if it has.)
• Consider asking first in a Discussion.

Motivation

Extend recommendation system to process and recommend content containing both text and images.

Solution

• Use multi-modal embedding models (e.g., CLIP) to encode both textual and visual content into a shared feature space.
• Adjust recommendation algorithms to consider both modalities when ranking posts.
• Update recsys.py to load and utilize multi-modal models where applicable.

Alternatives

No response

Additional context

No response

[Ternura143]

Hi @Yuujee, I can take issue #143 and implement multimodal embeddings for the RecSys, updating recsys.py accordingly.
Could you assign it to me, and who should I coordinate with for the expected design and requirements?