BUG: Ineffecient way of getting images

### Is there an existing issue for this?

- [x] I have searched the existing issues

### What happened?

**Where** : `backend/app/database/images.py`

**What happened ?** : `db_get_all_images` fetches raw rows (potentially multiple per image due to joins) and iterates in Python to group tags into `images_dict`.

**What should be** : Use `GROUP_CONCAT` in the SQL query to aggregate tags per image ID in the database layer, returning one row per image.

**Why?**: Python iteration over 100k+ rows is significantly slower than C-based SQLite aggregation.

### Record

- [x] I agree to follow this project's Code of Conduct