test(mcp): analyze the seeded rows before the metric-series golden test

The first case spent 14 s per full-snapshot build in the all-time pulse
aggregate, planned against a table with no statistics, until autovacuum
analyzed it; on a loaded runner that took it past its 60 s budget. Analyze
measurements and rollups after seeding, as a production table would be.
The cases and the comparison they make are unchanged.

tests/integration/mcp-metric-series-source-snapshot.test.ts

@@ -218,6 +218,13 @@ beforeAll(async () => {
   // The mood feature warms its rollup tier fire-and-forget on first read;
   // settle it now so both builds read the same tier.
   await ensureUserMoodRollupsFresh(userId);
+  // Planner statistics for the rows just written. Without them the
+  // all-time pulse aggregate (`readAllTimeExtremes`, day-weighted for pulse)
+  // plans against an empty table and runs for ~14 s per call until
+  // autovacuum's analyze happens to land, which on a loaded runner took the
+  // first case past its 60 s budget. A production table always has
+  // statistics, so this measures what production runs.
+  await prisma.$executeRawUnsafe("ANALYZE measurements, measurement_rollups");
 }, 120_000);
 
 describe("get_metric_series built from one source's rows", () => {