Use manual configuration for test_reduction

The example relies on a single block being launched, but a forall isn't guaranteed to generate a configuration with a single block - for example, on the RTX A6000 it creates the configuration `[2, 768]`. Explicitly configuring the launch resolves this issue.
sklam · May 23, 2022 · Mar 11, 2022 · Mar 16, 2022 · Apr 1, 2022 · Apr 5, 2022
commit 1b77d364962d77976f4c0c6c8311274e47e3c6e1
@@ -64,9 +64,9 @@ def array_sum(data):
         # ex_reduction.kernel.end
 
         # ex_reduction.launch.begin
-        array_sum.forall(len(a))(a)
-        print(a[0])  # 523776
-        sum(np.arange(1024))  # 523776
+        array_sum[1, nelem](a)
+        print(a[0])                  # 523776
+        print(sum(np.arange(1024)))  # 523776
         # ex_reduction.launch.end
 
         np.testing.assert_equal(a[0], sum(np.arange(1024)))