Convert TransferFunction's shallow copy to deep copy. #130

I spent quite a bit of time trying to debug an error and finally stumbled on the documentation:

# Beware: this is a shallow copy! This should be okay,
# but be careful.

It took me a bit of digging and debugging before I realized my errors caused by the shallow copy:

>>> print(num)
>>> print(type(num[0]))

[[1], [2]]
<class 'list'>

>>> sys1 = control.TransferFunction([num],[den])
>>> sys1

>>> print(num)
>>> print(type(num[0]))

[array([1]), array([2])]
<class 'numpy.ndarray'>

This can make scripts behave different based on the order of operation:

num = ...
den = ...
myfunc(num)
TransferFunction(num, den)

vs.

num = ...
den = ...
TransferFunction(num, den)
myfunc(num)

To new Python users having the variables 'num' & 'den' behave different before and after calling TransferFunction can be confusing and off-putting.