Nick/matlab support

[ntjohnson1]

Adds a matlab_support category since we are enabling a printing functionality to more closely match MATLAB I expect other features (or at least printing of all our classes). This uses the core functionality that Tammy wrote from the linked issue with some reformating.

This branches off #359 so that should merge first.
Resolves: #350

>>> import numpy as np
>>> import pyttb as ttb
>>> from pyttb.matlab.matlab_support import matlab_print
>>> tensor= ttb.tensor(np.arange(16), shape=(2,2,2,2))
>>> print(tensor)
tensor of shape (2, 2, 2, 2) with order F
data[:, :, 0, 0] =
[[0 2]
 [1 3]]
data[:, :, 1, 0] =
[[4 6]
 [5 7]]
data[:, :, 0, 1] =
[[ 8 10]
 [ 9 11]]
data[:, :, 1, 1] =
[[12 14]
 [13 15]]
>>> matlab_print(tensor)
This is a tensor of shape 2 x 2 x 2 x 2
	data(:,:, 0, 0) =
		0 2
		1 3
	data(:,:, 1, 0) =
		4 6
		5 7
	data(:,:, 0, 1) =
		8 10
		9 11
	data(:,:, 1, 1) =
		12 14
		13 15
>>>matlab_print(tensor[:,:,0,0])
This is a tensor of shape 2 x 2
	data(:,:) =
		0 2
		1 3
>>>matlab_print(tensor[:,0,0,0])
This is a tensor of shape 2
	data(:) =
		0
		1
>>>matlab_print(tensor[:,0,0,0], format="5.1f", name="X")
X is a tensor of shape 2
	X(:) =
		  0.0
		  1.0
# Also support printing the underlying numpy arrays
>>>matlab_print(tensor.data)
ndarray(:,:, 0, 0) =
	0 2
	1 3
ndarray(:,:, 1, 0) =
	4 6
	5 7
ndarray(:,:, 0, 1) =
	8 10
	9 11
>>>matlab_print(tensor.data, format="5.1f", name="X")
X(:,:, 0, 0) =
	  0.0   2.0
	  1.0   3.0
X(:,:, 1, 0) =
	  4.0   6.0
	  5.0   7.0
X(:,:, 0, 1) =
	  8.0  10.0
	  9.0  11.0
X(:,:, 1, 1) =
	 12.0  14.0
	 13.0  15.0

[tgkolda]

There is a huge indent at each level. Can that be lessened to something smaller and/or customizable?

[ntjohnson1]

The indent size may have been a copy paste issue. I followed the format here where each level is a single tab character.

We can set the indentation at each level to anything. You should be able to checkout my branch to try it, and the indentation level is set via the textwrap calls in matlab_utilities.