audit: store lock reads and region/area set resolution vs upstream PRs 2246 and 2247 #597

Audit task. Compare our store with two upstream fixes (Kpa-clawbot/CoreScope pull request numbers 2246 and 2247; literal references, not linked):

• 2246: read StoreTx state under s.mu in grouped packets and other readers;
• 2247: resolve region and area sets before taking s.mu.

We have related fixes (#475 races, #495 no SQL under s.mu, #583 SQL out of mutexes). Determine whether any reader still reads StoreTx state without the lock, or resolves region/area sets (possibly SQL) while holding s.mu. Use -race stress tests on every reader. Report the findings, and fix any gaps with regression tests; if there are none, report that with evidence.