Pluggable embeddings for self-hosted

Self-hosted supermemory can now use local or remote embedding models.

• Pluggable embeddings — local ONNX (default) or OpenAI / OpenAI-compatible / Google via SUPERMEMORY_EMBEDDING_*; first-boot picker + plan lock in embedding-plan.json
• Safe upgrades — populated stores without a lock assume legacy local · Xenova/bge-base-en-v1.5 · 768d; same-dimension model switches fail fast instead of mixing vectors

curl -fsSL https://supermemory.ai/install | bash

supermemory-server upgrade