### Changes:
- Added a new `scalingStrategy.md` document explaining limitations and solutions for handling large-scale data with HNSW:
- Highlighted in-memory limitations of standard HNSW implementations.
- Detailed disk-based, distributed, and hybrid approaches for scaling to terabyte-scale datasets.
- Provided real-world examples from systems like Qdrant, Milvus, FAISS, and DiskANN.
- Emphasized trade-offs between memory usage, search accuracy, and performance when scaling.
### Purpose:
Documented approaches for managing large-scale data with HNSW to assist developers in implementing efficient, scalable solutions. The guide serves as a reference for overcoming memory limitations and adopting advanced techniques for handling terabyte-scale datasets.
