Convert gravmag.prism Cython code to numpy

[leouieda]

Following what was done for polyprism in #368 and sphere in #364.
Also implements a regression test using saved data from a simple 1 prism model.

This is the benchmark script used:

from __future__ import division, print_function
import sys
import numpy as np
from fatiando import gridder, utils
from fatiando.mesher import PrismMesh
from fatiando.gravmag import prism
# Get processor information
tmp = !cat /proc/cpuinfo | grep "model name"
processor = tmp[0].split(':')[1].strip()
print(processor)
# Make a model for testing
model = PrismMesh((0, 1000, 0, 2000, 0, 500), (5, 3, 8))
model.addprop('density', 1000*np.ones(model.size))
model.addprop('magnetization', utils.ang2vec(2*np.ones(model.size), 25, -10))
inc, dec = -30, 50
x, y, z = gridder.regular((-500, 500, -500, 500), (50, 50), z=-1)
print('Model size: {}'.format(model.size))
print('Grid size: {}'.format(x.size))
# Time the forward modeling of gravity, gradients and mag
print('Times:')
if len(sys.argv) == 1:
    fields = 'potential gx gy gz gxx gxy gxz gyy gyz gzz tf bz bx by'.split()
else:
    fields = sys.argv[1:]
for field in fields:
    print('  {}: '.format(field), end='')
    if field == 'tf':
        args = (x, y, z, model, inc, dec)
    else:
        args = (x, y, z, model)
    %timeit -n 10 getattr(prism, field)(*args)

These are the results for the master branch:

Intel(R) Core(TM) i5-4200U CPU @ 1.60GHz
Model size: 120
Grid size: 2500
Times:
  potential: 10 loops, best of 3: 776 ms per loop
  gx: 10 loops, best of 3: 435 ms per loop
  gy: 10 loops, best of 3: 443 ms per loop
  gz: 10 loops, best of 3: 420 ms per loop
  gxx: 10 loops, best of 3: 201 ms per loop
  gxy: 10 loops, best of 3: 198 ms per loop
  gxz: 10 loops, best of 3: 220 ms per loop
  gyy: 10 loops, best of 3: 228 ms per loop
  gyz: 10 loops, best of 3: 235 ms per loop
  gzz: 10 loops, best of 3: 221 ms per loop
  tf: 10 loops, best of 3: 769 ms per loop
  bz: 10 loops, best of 3: 408 ms per loop
  bx: 10 loops, best of 3: 420 ms per loop
  by: 10 loops, best of 3: 415 ms per loop

Checklist:

• Make tests for new code (at least 80% coverage)
• Create/update docstrings
• Include relevant equations and citations in docstrings
• Docstrings follow the style conventions
• Code follows PEP8 style conventions
• Code and docs have been spellchecked
• Include new dependencies in doc/install.rst, environment.yml, ci/requirements.txt.
• Documentation builds properly (run cd doc; make locally)
• Changelog entry (leave for last to avoid conflicts)

[MohAbdrabou]

You can provide me script for calculating gravity due to 3d prism using Plouff or Nagy formula