fix(grid-rs): bump memory limit to 2 Gi

1 Gi OOM'd during a real chain step. Each forecast hour loads both a
surface and pressure-level HRRR GRIB2 (~40 MB combined), decodes via
wgrib2 subprocess (which allocates its own buffers), scores 92k grid
points × 23 bands (~2.1M intermediates), and writes 23 score files.
2 Gi absorbs the spike; still ~3× smaller than the Elixir-per-pod
budget we're replacing.

k8s/deployment-grid-rs.yaml

@@ -71,14 +71,14 @@ spec:
           resources:
             requests:
               cpu: 100m
-              memory: 256Mi
+              memory: 512Mi
             limits:
               cpu: "4"
-              # 512 Mi OOM-killed the real binary at startup. 1 Gi gives
-              # room for sqlx pool + reqwest pool + tokio runtime +
-              # tracing buffers, with headroom vs the steady-state
-              # target (~500 Mi during a chain step).
-              memory: 1Gi
+              # OOM at 1 Gi during active chain work (92k points × 23
+              # bands ≈ 2.1M scores, plus decoded GRIB2 buffers, plus
+              # wgrib2 subprocess memory). 2 Gi leaves headroom; we're
+              # still an order of magnitude below the 6 Gi Elixir budget.
+              memory: 2Gi
           securityContext:
             allowPrivilegeEscalation: false
             runAsNonRoot: true