Fix commit and query contention

[jewei1997]

Describe your changes and provide context

Previously, there was an issue where Commit() and queries (getLogs, getReceiptByTxHash) into parquet via DuckDB would sometimes have lock contention because these queries would hold the read lock for the entire query, while Commit() stores receipts and during parquet file rotation, we need to acquire this lock to update the files on disk we are tracking. This has the possibility of slowing down the consensus commit phase, potentially slowing down block production.

The fix: old hold the existing lock for modifying and reading/copying of the files being tracked. This allows for minimal contention between Commit() and queries(). Then introduce a second lock which will synchronize between queries and pruning so that parquet files aren't dropped while queries are happening. Pruning and queries are less performance-critical to the node as ensuring Commits are fast.

Testing performed to validate your change

unit tests.

[github-actions]

The latest Buf updates on your PR. Results from workflow Buf / buf (pull_request).

Build	Format	Lint	Breaking	Updated (UTC)
✅ passed	✅ passed	✅ passed	✅ passed	Mar 17, 2026, 1:20 PM

[codecov]

Codecov Report

❌ Patch coverage is 91.66667% with 3 lines in your changes missing coverage. Please review.
✅ Project coverage is 58.42%. Comparing base (dcdde1b) to head (2486772).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
sei-db/ledger_db/parquet/store.go	80.00%	1 Missing and 1 partial ⚠️
sei-db/ledger_db/parquet/reader.go	96.15%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3075      +/-   ##
==========================================
- Coverage   58.43%   58.42%   -0.02%     
==========================================
  Files        2088     2088              
  Lines      172084   172102      +18     
==========================================
- Hits       100559   100547      -12     
- Misses      62599    62616      +17     
- Partials     8926     8939      +13     

Flag	Coverage Δ
sei-chain-pr	72.96% <91.66%> (?)
sei-db	70.41% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
sei-db/ledger_db/parquet/reader.go	77.51% <96.15%> (+1.99%)	⬆️
sei-db/ledger_db/parquet/store.go	68.45% <80.00%> (+0.27%)	⬆️

... and 13 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[masih]

Instead of passing in t to then call require no error, I recommend using an error group and have this function return error.

This way we would capture errors in the goroutine. Otherwise t would exist the current goroutine only and might not fail the test itself on error.