v0.40.0-rc6: model: add multimodal embeddings (#18820)

发生了什么

Implements the EmbeddingGemma2Model architecture on the MLX runner: 24-layer bidirectional text encoder with PLE, shared gemma4 vision/audio towers, mean-pool + L2 output. /api/embed accepts per-item media via input dicts.

摘要按规则整理自下方来源原文

来源