Debug Trace Speedup

[Kbhat1]

Describe your changes and provide context

• Avoid store-tracer overhead on ordinary debug_trace* calls; use debug_traceStateAccess when Cosmos/store-level access tracing is needed
• Keep profiled block tracing on the default tracer path while preserving legacy semantics for explicit tracers like flatCallTracer
• Parallelize native/default block trace execution across per-tx prestate snapshots while preserving ordered state advancement (largest improvement)

Testing performed to validate your change

• Tested on node

[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	❌ failed (0)	Mar 18, 2026, 8:54 PM

[codecov]

Codecov Report

❌ Patch coverage is 46.72897% with 171 lines in your changes missing coverage. Please review.
✅ Project coverage is 58.18%. Comparing base (8a11ce5) to head (896d58d).
⚠️ Report is 2 commits behind head on main.

Files with missing lines	Patch %	Lines
evmrpc/block_trace_profiled.go	45.19%	124 Missing and 30 partials ⚠️
evmrpc/tracers.go	72.72%	2 Missing and 4 partials ⚠️
x/evm/state/balance.go	25.00%	2 Missing and 4 partials ⚠️
sei-cosmos/types/context.go	42.85%	2 Missing and 2 partials ⚠️
x/evm/state/code.go	0.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3058      +/-   ##
==========================================
- Coverage   58.61%   58.18%   -0.43%     
==========================================
  Files        2090     2074      -16     
  Lines      172771   171777     -994     
==========================================
- Hits       101262    99946    -1316     
- Misses      62532    62878     +346     
+ Partials     8977     8953      -24     

Flag	Coverage Δ
sei-chain-pr	68.73% <46.72%> (?)
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
app/app.go	61.13% <100.00%> (-6.09%)	⬇️
x/evm/state/code.go	82.35% <0.00%> (ø)
sei-cosmos/types/context.go	92.30% <42.85%> (-1.37%)	⬇️
evmrpc/tracers.go	65.19% <72.72%> (-0.37%)	⬇️
x/evm/state/balance.go	60.20% <25.00%> (-1.76%)	⬇️
evmrpc/block_trace_profiled.go	45.19% <45.19%> (ø)

... and 98 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.

[codchen]

correctness-wise I don't see issues when eyeballing the change, but we should run blockaudit against 10k blocks within each historical version if not done already, given the abundance of edge cases in those older upgrades.
performance-wise, if I'm understanding correctly, the parallel path would execute each transaction twice: once to build prestate (which is sequential, without logger), and once to generate traces (which is parallel and with logger). This makes sense if state access is no longer the bottleneck and the logger now contributes the most to latency, but if this condition changes in the future (e.g. maybe with longer history, state access slows down again) then we might need to revisit.