Speed up ellipsoid image creation

[mdoube]

Describe the bug Ellipsoid image creation takes about twice as long in EF2 as in EF1

Expected behavior EF2 should be faster than EF1

Additional context Each foreground pixel has to be given the EF value of the maximal ellipsoid that contains it.

In EF1 that was achieved by limiting search scope aggressively:

• the z bound (and y bound) of each ellipsoid was calculated so that a shortlist of potential ellipsoids was generated for each slice/row
• the ellipsoid list was sorted once and traversed from largest to smallest, so that the whole list did not need to be searched each time: the first positive contains() hit was guaranteed to be the biggest, no need to search the rest.
• the contains() method was optimised to exclude unlikely ellipsoids cheaply and early in its execution

EF2 does this:

	private Img<IntType> assignEllipsoidIDs(final Img<BitType> mask, final List<QuickEllipsoid> ellipsoids) {

		final Img<IntType> idImage = ArrayImgs.ints(mask.dimension(0), mask.dimension(1),weightedAverageN, mask.dimension(2));
		idImage.forEach(c -> c.setInteger(-1));

		for(int nn=0;nn<weightedAverageN;nn++) {
			final int n = nn;
			final Map<QuickEllipsoid, Integer> iDs = IntStream.range(0, ellipsoids.size()).boxed()
					.collect(toMap(ellipsoids::get, Function.identity()));
			final LongStream zRange = LongStream.range(0, mask.dimension(2));
			zRange.parallel().forEach(z -> {
				// multiply by image unit? make more intelligent bounding box?
				final List<QuickEllipsoid> localEllipsoids = ellipsoids.stream()
						.filter(e -> Math.abs(e.getCentre()[2] - z) < e.getSortedRadii()[2]).collect(toList());
				final long[] mins = {0, 0, z};
				final long[] maxs = {mask.dimension(0) - 1, mask.dimension(1) - 1, z};
				final Cursor<BitType> maskSlice = Views.interval(mask, mins, maxs).localizingCursor();
				colourSlice(idImage, maskSlice, localEllipsoids, iDs, n);
			});
		}
		return idImage;
	}