[Feature Request] Integrate GoodMem as a memory toolkit

### Required prerequisites - [x] I have searched the [Issue Tracker](https://github.com/camel-ai/camel/issues) and [Discussions](https://github.com/camel-ai/camel/discussions) that this hasn't already been reported. (+1 or comment there if it has.) - [ ] Consider asking first in a [Discussion](https://github.com/camel-ai/camel/discussions/new). ### Motivation [GoodMem](https://goodmem.ai/) is a self-hosted memory infrastructure for AI agents. It handles embedding, chunking, storage, and retrieval on the server, so users can store and search content semantically without building their own pipelines. Capabilities that would benefit CAMEL users: - Bring your own embedding model: supports OpenAI, Voyage AI, Cohere, vLLM, TEI, and Llama.cpp, including fully local and offline models - Hybrid search: combine dense and sparse embedders (e.g., MiniLM + SPLADE) in a single space with configurable weights for both semantic and keyword-match queries - Configurable chunking: chunk size, overlap, separators, and mode are set per space and handled entirely on the server - File ingestion: upload PDFs, DOCX, images, and other formats without manual text extraction - Metadata filtering: SQL-style filters with JSONPath extraction, date ranges, regex, array membership, and nested object traversal - Reranking: pluggable reranker models re-score results after retrieval for better relevance - Auto-summary: an LLM generates a consolidated answer from retrieved chunks at query time, returned alongside the raw results - Deep Research mode: runs multiple iterative search rounds, refining the query between rounds to handle complex or open-ended topics - Agent-driven memory management: exposes tools that let agents autonomously create spaces, store documents, and manage memory lifecycle, not just search - Self-hosted, no usage caps: no artificial limits on documents, spaces, or requests This gives CAMEL users a self-hosted, batteries-included option for agent memory. ### Solution A `GoodMemToolkit` extending `BaseToolkit` that wraps the GoodMem API and exposes operations as agent-callable tools: managing spaces, storing memories (text or files), semantic retrieval with streaming, and memory lookup/deletion. Also supports MCP server capability. A working implementation with unit tests is ready for review. Documentation: https://docs.goodmem.ai/docs/ ### Alternatives _No response_ ### Additional context _No response_