[Feature] Lifecycle support for storing data into S3

## Summary
Add lifecycle support for storing data into S3, enabling tiered/archival storage of BanyanDB data.

## Task Requirements
1. **Research** — survey potential solutions (S3 storage backends, lifecycle policies, integration approaches with BanyanDB's existing storage model).
2. **PoC** — implement a proof-of-concept for the chosen approach.
3. **Benchmark** — run benchmarks (write/read latency, throughput, cost) before finalizing the design.

## Acceptance Criteria
- Research document comparing candidate solutions.
- Working PoC.
- Benchmark results.
- Final design proposal based on PoC + bench findings.

Start by surveying S3 storage backends, lifecycle policies, and integration approaches with BanyanDB's existing storage model. Produce a comparison document, build a working proof-of-concept, run write/read latency, throughput, and cost benchmarks, and use the findings to complete a final design proposal.

Written by the indexing model from the issue text.