v0.5.5 - Improved cache, batch processing, health endpoint

Added

• Improved cache by implementing LRU with a max cache size
• Batch processing of multiple texts to embed with single request
• /health endpoint to return server status information

API server with cache

• Updated API endpoint with more verbose responses including time taken to generate an embedding, the model used, and status
• In-memory cache for embeddings
• Python client for API interaction
• Better code organization

The release tag is completely arbitrary at this point.