Update default value of mrd_pool_size in micro benchmark to None and add pool parameters to benchmark schema

[Yonghui-Lee]

• The read microbenchmarks were previously defaulting mrd_pool_size to 1. It will overrode the environment's DEFAULT_GCSFS_CONCURRENCY and forced the use of only a single gRPC stream. I have changed the default to None, allowing it to correctly inherit environment concurrency settings and restore multi-stream gRPC read performance with the default config.
• Updates the BigQuery staging schema definition to declare both mrd_pool_cache_size and mrd_pool_size as STRING fields. This allows the ingestion pipeline to gracefully parse new results.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 89.13%. Comparing base (afcf7a4) to head (2da24d3).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #850   +/-   ##
=======================================
  Coverage   89.13%   89.13%           
=======================================
  Files          15       15           
  Lines        3130     3130           
=======================================
  Hits         2790     2790           
  Misses        340      340           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.