test: fix CI red on #3130 — withEnv for batch-concurrency env + close resetGateway shard-order poison window

Two real failures surfaced by this PR's re-sharding:

1. verify/check-test-isolation: embed-batch-concurrency.test.ts mutated
   process.env directly (R1). Now uses withEnv().

2. test (9) source-health "expected 1280 dimensions, not 1536": a file
   whose last afterEach calls resetGateway() leaves the gateway slot
   empty during the NEXT file's beforeAll (which runs before any
   beforeEach can restore the legacy 1536 pin), so initSchema() sizes
   the embedding column from the zembed-1/1280 defaults and every
   1536-d fixture in that file fails. Which pair collides depends on
   shard composition, so adding test files (as this PR does) can
   surface it anywhere. The legacy-embedding preload now also repairs
   the empty slot in a global afterEach (preload after-hooks run after
   file-local ones), closing the window at the root instead of
   patching one victim file.

Reproduced locally with a poison/afterEach-reset file followed by a
schema-creating file: embedding column typmod 1280 before the fix,
1536 after. check-test-isolation, typecheck, and the affected suites
all pass.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

test/embed-batch-concurrency.test.ts

@@ -24,6 +24,7 @@ import {
   __setEmbedTransportForTests,
 } from '../src/core/ai/gateway.ts';
 import { embedBatch } from '../src/core/embedding.ts';
+import { withEnv } from './helpers/with-env.ts';
 
 const DIMS = 1536;
 
@@ -69,7 +70,6 @@ describe('embedBatch bounded parallelism (#1818)', () => {
   });
   afterEach(() => {
     __setEmbedTransportForTests(null);
-    delete process.env.GBRAIN_EMBED_BATCH_CONCURRENCY;
   });
 
   test('default pool dispatches sub-batches in parallel, order preserved', async () => {
@@ -92,9 +92,10 @@ describe('embedBatch bounded parallelism (#1818)', () => {
   });
 
   test('GBRAIN_EMBED_BATCH_CONCURRENCY env bounds the pool when option unset', async () => {
-    process.env.GBRAIN_EMBED_BATCH_CONCURRENCY = '2';
     const tracker = installTrackingTransport();
-    await embedBatch(texts, { onBatchComplete: () => {} });
+    await withEnv({ GBRAIN_EMBED_BATCH_CONCURRENCY: '2' }, async () => {
+      await embedBatch(texts, { onBatchComplete: () => {} });
+    });
     expect(tracker.maxInFlight()).toBeGreaterThan(1);
     expect(tracker.maxInFlight()).toBeLessThanOrEqual(2);
   });

test/helpers/legacy-embedding-preload.ts

@@ -19,7 +19,7 @@
  * overwrites this preload.
  */
 import { configureGateway, getEmbeddingDimensions } from '../../src/core/ai/gateway.ts';
-import { beforeEach } from 'bun:test';
+import { afterEach, beforeEach } from 'bun:test';
 
 const LEGACY_CONFIG = {
   embedding_model: 'openai:text-embedding-3-large',
@@ -52,7 +52,7 @@ applyLegacy();
 //   2. file-local beforeAll → may overwrite to ZE/1280
 // Since beforeAll runs once per file BEFORE the first beforeEach,
 // file-local beforeAll wins for that file's tests. ✓
-beforeEach(() => {
+function applyLegacyIfEmpty() {
   try {
     // Only re-apply if the gateway was reset (or never configured).
     // Tests that explicitly configured a different model in their
@@ -62,4 +62,28 @@ beforeEach(() => {
   } catch {
     applyLegacy();
   }
-});
+}
+
+beforeEach(applyLegacyIfEmpty);
+
+// PR #3130 shard-order fix: beforeEach alone leaves ONE window open — a file
+// whose LAST afterEach calls resetGateway() poisons the NEXT file's
+// beforeAll, which runs BEFORE any beforeEach fires. A beforeAll there that
+// does engine.initSchema() then sizes the embedding column from the gateway
+// DEFAULTS (zembed-1/1280d) instead of the pinned legacy 1536, and every
+// 1536-d Float32Array fixture in that file dies with
+// "expected 1280 dimensions, not 1536". Which file pair collides is a
+// function of shard composition, so adding/removing ANY test file can
+// surface it (that is exactly how it bit shard 9).
+//
+// Preload hooks are registered before any file-local hooks, and bun runs
+// after-hooks inside-out (file-local afterEach first, then this one), so
+// this repairs the empty slot immediately after the poisoning reset —
+// before the next file's beforeAll can observe it.
+//
+// Known remaining window: a file whose afterAll() resets the gateway (no
+// hook runs between its afterAll and the next file's beforeAll). Files
+// that reset in afterAll and can precede a schema-creating file should
+// re-apply their own config, or the victim file should configureGateway()
+// explicitly in its beforeAll.
+afterEach(applyLegacyIfEmpty);