fix(settings): stop normalizeOpenItemCap's cap leaking into unrelated fields

[JSONbored]

Summary

• PR fix(settings): bound contributor open caps #3977 correctly clamped normalizeOpenItemCap to MAX_CONTRIBUTOR_OPEN_ITEM_CAP (100) for contributorOpenPrCap/contributorOpenIssueCap, since live enforcement only ever samples a fixed 100-row budget.
• That function is also reused by three unrelated settings purely for its "positive integer or null" validation, and they silently inherited the same 100 ceiling:
  • normalizeModerationDecayDays (violation-decay days, intended ceiling MAX_MODERATION_VIOLATION_DECAY_DAYS = 3650) — double-clamped down to 100 first, so its own ceiling was never reachable. Caught by 3 already-failing assertions in test/unit/moderation-config-db.test.ts (e.g. expected 100 to be 3650).
  • requireFreshRebaseWindowMinutes and accountAgeThresholdDays — silently capped at 100 (minutes / days respectively) with no existing test exercising a value above 100, so this was live but latent.
• Extracts the shared "positive integer or null" validation into normalizePositiveIntOrNull (no cap). normalizeOpenItemCap and normalizeModerationDecayDays each apply their own, unrelated ceiling on top of it; the other two fields use the uncapped helper directly.

Scope

• src/db/repositories.ts (the shared helper + its 5 call sites) and test/unit/data-spine.test.ts (new regression coverage for the 2 latent fields).

Validation

• npx tsc --noEmit — clean.
• npx vitest run test/unit/data-spine.test.ts test/unit/moderation-config-db.test.ts — 35/35 pass (the 3 previously-failing moderation assertions now pass without being touched).
• Scoped coverage on every changed line range in src/db/repositories.ts: 100%.

Safety

• No secrets/wallets/hotkeys/trust-scores/reward-values touched.
• No changelog edit; no site//CNAME/lovable changes.

[superagent-security]

Superagent didn't find any vulnerabilities or security issues in this PR.

[codecov]

❌ 1 Tests Failed:

Tests completed	Failed	Passed	Skipped
11138	1	11137	7

View the full list of 1 ❄️ flaky test(s)

test/unit/queue.test.ts > queue processors > a #1960 action-command verb with no dispatch handler wired yet (e.g. pause) is bailed out of the Q&A answer-card path, not misrendered as help (#2160)

Flake rate in main: 100.00% (Passed 0 times, Failed 1 times)

Stack Traces | 0.872s run time

AssertionError: expected 1 to be +0 // Object.is equality

- Expected
+ Received

- 0
+ 1

 ❯ test/unit/queue.test.ts:22807:28

To view more test analytics, go to the Test Analytics Dashboard
_{📋 Got 3 mins? Take this short survey to help us improve Test Analytics.}