Fix flaky/environment-dependent tests (macOS)

[shemnon]

Summary

• Stop double-logging recovered panics in evmrpc once a panic hook is set (the hook currently has a single caller, a test, that only needs the notification and never depended on the print/stack trace also firing)
• Fix TestFreezeModeDisablesMempoolTraffic flaking on macOS: it reserved a listen address via TestReserveAddr and handed it to node.Start, but net.Listen can't adopt that reservation and the freed :0 bind is stealable by another process in the window before Start runs; the test only calls rpcEnv in-process, so skip the TCP listener entirely
• Fix a timezone-dependent flake in the mint-release test: NewMinter derived start/end dates from local wall-clock time instead of UTC calendar dates, and MintRelease re-derived the expected release amount by hand instead of via Minter.GetReleaseAmountToday

Test plan

• go test ./evmrpc/...
• go test ./sei-tendermint/node/...
• go test ./testutil/processblock/...

[cursor]

PR Summary

Low Risk
Changes are limited to test helpers, one tendermint test config, and evmrpc panic logging order with no production behavior change beyond optional hook-only logging.

Overview
Hardens test and recovery behavior so CI is less sensitive to timezone and listen-address races on macOS, and panic hooks behave as documented.

In evmrpc, recovered panics now only invoke SetPanicHook when one is registered—the default fmt.Printf + stack trace run only if no hook handles the panic (documented as replacing default logging).

Mint processblock tests build minter start/end from UTC midnight calendar dates instead of local time.Now(), and MintRelease asserts expected mint via Minter.GetReleaseAmountToday rather than hand-parsed date span math (mirrored under giga/deps/testutil).

TestFreezeModeDisablesMempoolTraffic sets cfg.RPC.ListenAddress to empty so the node does not bind TCP; the test exercises rpcEnv in-process and avoids the flaky TestReserveAddr / stealable :0 window on macOS.

^{Reviewed by Cursor Bugbot for commit afac58d. Bugbot is set up for automated code reviews on this repo. Configure here.}

[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	Sep 9, 2026, 9:15 PM

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 60.25%. Comparing base (c78d1c5) to head (afac58d).
⚠️ Report is 7 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #4115      +/-   ##
==========================================
- Coverage   61.21%   60.25%   -0.97%     
==========================================
  Files        2196     2087     -109     
  Lines      193436   180445   -12991     
==========================================
- Hits       118416   108719    -9697     
+ Misses      63790    61580    -2210     
+ Partials    11230    10146    -1084     

Flag	Coverage Δ
sei-chain-pr	71.14% <100.00%> (?)
sei-db	69.80% <ø> (ø)
sei-db-state-db	?

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

Files with missing lines	Coverage Δ
evmrpc/utils.go	72.81% <100.00%> (+0.12%)	⬆️
testutil/processblock/genesismint.go	100.00% <100.00%> (ø)
testutil/processblock/verify/mint.go	100.00% <100.00%> (+18.18%)	⬆️

... and 110 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.

[seidroid]

Three small, well-targeted flake fixes: the evmrpc panic-hook change keeps its only test passing unchanged and leaves production logging intact (no hook is set outside tests), and dropping the RPC listener from TestFreezeModeDisablesMempoolTraffic is safe since every assertion goes through rpcEnv in-process. The mint fix is right in direction, but the new expected-amount computation uses the block time instead of the epoch start time the mint hook actually uses, leaving a narrow midnight-crossing window.

Findings: 0 blocking | 1 non-blocking | 1 posted inline

Blockers

• None at the file/PR level.

Non-blocking

• None at the file/PR level.
• 1 suggestion(s)/nit(s) flagged inline on specific lines.

[cursor]

Cursor Bugbot has reviewed your changes using default effort and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

Want higher recall? High effort reviews run extra passes and find more bugs. A team admin can switch effort levels in the Cursor dashboard.

^{Reviewed by Cursor Bugbot for commit 6cefa3e. Configure here.}