feat: initialize the memorize and retrieve workflows with the new 3-layer architecture

[ankaisen]

No description provided.

[Copilot]

Pull Request Overview

This PR initializes the memorize and retrieve workflow by implementing a new 3-layer architecture. The changes remove the old server and SDK implementations in favor of a streamlined core library structure focused on memory operations with direct LLM integration.

Key Changes:

• Removed legacy server, SDK, and old memory system code
• Introduced new core modules for memory operations (storage, vector indexing, LLM clients)
• Implemented Rust bindings for Python package
• Updated project structure to use maturin build system with Python 3.14+ requirement

Reviewed Changes

Copilot reviewed 128 out of 141 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
pyproject.toml	Migrated from setuptools to maturin build system, updated dependencies and Python version requirement to 3.14+
src/memu/app/service.py	New core service implementing 3-layer architecture for memorize and retrieve workflows
src/memu/llm/http_client.py	HTTP-based LLM client with support for multiple backends
src/memu/utils/video.py	Video frame extraction utilities using ffmpeg
src/memu/storage/local_fs.py	Local filesystem storage implementation for resources
src/memu/vector/index.py	Vector similarity search implementation using cosine distance
src/memu/prompts/*	Prompt templates for various memory operations
src/lib.rs	Rust entry point for Python bindings
tests/rust_entry_test.py	Basic test for Rust-Python integration

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.