BUG: vario_estimate alters field

[MuellerSeb]

Problem: gs.vario_estimate could alter the input field, when given as a numpy array with floats along with a trend function:

import numpy as np
import gstools as gs

trend = lambda a, b: a
x = y = t = np.array([1.0, 2.0, 3.0, 4.0])
gs.vario_estimate((x, y), t, trend=trend)
print(t)

results in

[0. 0. 0. 0.]

Although it should be [1. 2. 3. 4.]

To solve this, we need to set copy=True here:

GSTools/gstools/variogram/variogram.py

Line 246 in f91d3b1

field = np.ma.array(field, ndmin=2, dtype=np.double)