test: port upstream fixture and retry fixes (2a4b8f2)

The CLI suite is red on this branch for two reasons that upstream already
fixed and that never made it here.

parser.test.ts (a)/(f): createJsonlSession stamped fixture events at a fixed
2026-05-01. Durable providers age out at 90 days, so on 2026-07-30 the fixture
silently pruned to zero and both cases began failing with `expected +0 to be
200`. Confirmed directly: the identical fixture dated relative to now yields
one project and 200 output tokens; with the literal it yields none. Date the
events relative to now so they stay inside the window whenever the suite runs.

vitest.config.ts / cache-refresh-lock: a handful of integration tests exercise
real servers, spawned subprocesses and real filesystem locks, and starve under
a saturated parallel run — failing closed, which is correct behaviour but not
what those tests measure. A small global retry rides that out; a real
regression is deterministic and fails every attempt.

Two upstream hunks are deliberately not ported. cli-durable-totals already has
an equivalent fix on this branch (ec449af) that clamps to a fraction of the
elapsed day rather than a fixed offset. The `corrupt lock recovery` describe
does not exist here, because upstream's d514459 has not been ported either.

Test-only; no production code changed.

packages/cli/tests/cache-refresh-lock.test.ts

@@ -208,7 +208,7 @@ describe('warm session-cache refresh lock', () => {
   // run (fs 'unavailable' makes the fence fail CLOSED, which is correct but
   // not what this test measures); the actual race fails ~6% per verify, so a
   // mutated build cannot pass any attempt.
-  it('the fence never loses to its own heartbeat (in-process serialization)', { retry: 5 }, async () => {
+  it('the fence never loses to its own heartbeat (in-process serialization)', { retry: 10 }, async () => {
     // Regression: verifyStillOwner and the heartbeat tick both take the
     // takeover guard; without in-process serialization the fence could observe
     // its own heartbeat's guard file and abort a legitimate publication.

packages/cli/tests/parser.test.ts

@@ -143,10 +143,15 @@ async function createJsonlSession(
   const dir = join(sessionStateDir, sessionId)
   await mkdir(dir, { recursive: true })
   await writeFile(join(dir, 'workspace.yaml'), `id: ${sessionId}\ncwd: /home/user/testproj\n`)
+  // Dated relative to now so the session stays inside the 90-day retention
+  // window whenever the suite runs; a fixed literal silently ages out (these
+  // events were `2026-05-01`, which prunes to zero once now is 90 days past it).
+  const base = Date.now() - 2 * 24 * 60 * 60 * 1000
+  const ts = (offsetSec: number) => new Date(base + offsetSec * 1000).toISOString()
   const lines = [
-    JSON.stringify({ type: 'session.model_change', timestamp: '2026-05-01T10:00:00Z', data: { newModel: 'gpt-4.1' } }),
+    JSON.stringify({ type: 'session.model_change', timestamp: ts(0), data: { newModel: 'gpt-4.1' } }),
-    JSON.stringify({ type: 'user.message', timestamp: '2026-05-01T10:00:05Z', data: { content: 'hello', interactionId: 'int-1' } }),
+    JSON.stringify({ type: 'user.message', timestamp: ts(5), data: { content: 'hello', interactionId: 'int-1' } }),
-    JSON.stringify({ type: 'assistant.message', timestamp: '2026-05-01T10:00:10Z', data: { messageId: 'msg-1', outputTokens, interactionId: 'int-1', toolRequests: [] } }),
+    JSON.stringify({ type: 'assistant.message', timestamp: ts(10), data: { messageId: 'msg-1', outputTokens, interactionId: 'int-1', toolRequests: [] } }),
   ]
   await writeFile(join(dir, 'events.jsonl'), lines.join('\n') + '\n')
   return join(dir, 'events.jsonl')

packages/cli/vitest.config.ts

@@ -6,5 +6,13 @@ export default defineConfig({
     // session-discovery env vars (CLAUDE_CONFIG_DIRS, HOME, XDG_*, every
     // provider-specific *_HOME) don't bleed real local data into fixtures.
     setupFiles: ['./tests/setup/env-isolation.ts'],
+    // A handful of integration tests exercise real servers, spawned CLI
+    // subprocesses and real filesystem locks. Under a saturated full-suite run
+    // an fs/socket op can starve and the operation fails closed (correct, but
+    // an environmental blip, not a logic error), so a different one trips each
+    // run. A small retry rides out that starvation; a real regression is
+    // deterministic and fails every attempt. Tests that need more headroom set
+    // a higher retry locally (it overrides this).
+    retry: 2,
   },
 })