Parallelize pixelize_sph_kernel_* methods. Fixes yt-project#2682

yt/utilities/lib/pixelization_routines.pyx

@@ -1032,7 +1032,7 @@ def pixelize_sph_kernel_projection(
                 prefactor_j *= quantity_to_smooth[j] * _weight_field[j]
 
             # found pixels we deposit on, loop through those pixels
-            for xi in range(x0, x1):
+            for xi in prange(x0, x1, schedule="static"):
                 # we use the centre of the pixel to calculate contribution
                 x = (xi + 0.5) * dx + x_min
 
@@ -1279,7 +1279,7 @@ def pixelize_sph_kernel_slice(
 
             # Now we know which pixels to deposit onto for this particle,
             # so loop over them and add this particle's contribution
-            for xi in range(x0, x1):
+            for xi in prange(x0, x1, schedule="static"):
                 x = (xi + 0.5) * dx + x_min
 
                 posx_diff = posx[j] - x
@@ -1371,7 +1371,7 @@ def pixelize_sph_kernel_arbitrary_grid(np.float64_t[:, :, :] buff,
 
             # Now we know which voxels to deposit onto for this particle,
             # so loop over them and add this particle's contribution
-            for xi in range(x0, x1):
+            for xi in range(x0, x1, schedule="static"):
                 x = (xi + 0.5) * dx + x_min
 
                 posx_diff = posx[j] - x