Add tier1 SQLite DB reuse

[SonAIengine]

Summary

• add persistent SQLite DB options to the tier1 benchmark runner
• write and validate .tier1.json sidecar metadata before reusing a benchmark DB
• add overwrite cleanup for SQLite/WAL/HNSW/metadata artifacts
• document persistent MS MARCO 1M repeat-run commands

Why

The MS MARCO 1M run proved the path works, but build/materialization took 1913.3s (31.9m). Reusing a validated persistent DB lets follow-up quality/search runs skip that cost.

Tests

• uv run --extra dev ruff check examples/ablation/run_tier1_benchmarks.py tests/test_tier1_benchmarks.py
• uv run --extra dev ruff format --check examples/ablation/run_tier1_benchmarks.py tests/test_tier1_benchmarks.py
• uv run --extra dev pytest tests/test_tier1_benchmarks.py -q
• PYTHONUNBUFFERED=1 uv run --extra sqlite python examples/ablation/run_tier1_benchmarks.py --only msmarco --subset 1 --corpus-limit 2 --use-sqlite-graph --sqlite-db-path /tmp/tier1_reuse_smoke.db --overwrite-sqlite-db --ingest-batch 1 --progress-every 1
• PYTHONUNBUFFERED=1 uv run --extra sqlite python examples/ablation/run_tier1_benchmarks.py --only msmarco --subset 1 --corpus-limit 2 --use-sqlite-graph --sqlite-db-path /tmp/tier1_reuse_smoke.db --reuse-sqlite-db --ingest-batch 1 --progress-every 1