Document MS MARCO 1M scale result

[SonAIengine]

Summary

• document the completed MS MARCO 1M large-tier run
• add ingest progress output to the tier1 runner for long corpus builds
• cover progress logging with a focused unit test

1M large-tier result

• MS MARCO passage dev, 1,000,000 docs, 50 queries, embedder-free SQLite/EvidenceSearch
• MRR@10 0.462, R@5 0.543, R@10 0.580, Hit@10 30/50
• Build 1913.3s (31.9m), Search 69.9s
• Local shard: 361 MB corpus JSONL + 511 KB manifest, gitignored

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 --ingest-batch 1 --progress-every 1
• PYTHONUNBUFFERED=1 uv run --extra sqlite python examples/ablation/run_tier1_benchmarks.py --only msmarco --subset 50 --corpus-limit 1000000 --use-sqlite-graph