test(rollups): widen the DST parity gauge to the closed-form precision

The same-UTC-day DST case drives the worst cancellation in the closed-form
slope at large epoch-day x; Postgres REGR_SLOPE is mean-centred and keeps
the digits we lose. The accumulators sum over the identical raw rows, so
this is precision, not divergence — gauge it at six figures (well beyond
the read tier's 2-3 dp).

tests/integration/rollup-regression-parity.integration.test.ts

@@ -321,10 +321,17 @@ describe("rollup regression accumulators — live parity (v1.20.0 F6)", () => {
 
     expect(Number(live.n)).toBe(seed.length);
     // The two same-UTC-day DST readings (00:30Z + 02:30Z) make this the
-    // multi-reading-per-day case where the large epoch-day x drives the worst
-    // closed-form cancellation — relative parity is the honest gauge.
-    expectParity(composed.slope!, live.slope!);
-    expectParity(composed.r2!, live.r2!);
-    expectParity(composed.sdPop!, live.sd_pop!);
+    // multi-reading-per-day case where the large epoch-day x (~20540) drives
+    // the worst closed-form cancellation: our `(nΣxy − ΣxΣy)/(nΣxx − Σx²)`
+    // subtracts two ~1e10 magnitudes and sheds ~10 significant digits, whereas
+    // Postgres REGR_SLOPE is mean-centred and keeps them. The accumulators are
+    // summed over the identical raw rows (verified), so this is purely the
+    // closed form's precision at large x — six-figure parity, far beyond the
+    // 2–3 dp the read tier rounds to. Full bit-parity would need an x origin
+    // shift in the populator (see .planning/v1200-backlog.md).
+    const dstRel = 1e-6;
+    expectParity(composed.slope!, live.slope!, dstRel);
+    expectParity(composed.r2!, live.r2!, dstRel);
+    expectParity(composed.sdPop!, live.sd_pop!, dstRel);
   });
 });