Overlapping of nodes in the draw diagram

[Ramshell]

Describe the bug
The draw function overlaps edges, making it seem that they share nodes.

To Reproduce
Steps to reproduce the behavior:

1. Run the following code

import matplotlib.pyplot as plt
import hypernetx as hnx
articles = {
        'CA': ('1', '3', '5', '6', '13', '80', '92', '8', '24', '36', '63', '75', '86', '88', '91',
                '10', '33', '57', '84', '27', '50', '67', '18', '60', '69', '81', '23', '31', '45', '87',
                '4', '71', '16', '17', '30', '55', '62', '93', '2', '19', '48', '7', '14', '15', '44', '58'),
        'MRDM': ('1', '3', '5', '6', '13', '27', '38', '42', '67', '80',
                 '92', '9', '23', '29', '31', '34', '35', '45', '87', '88',
                 '33', '84', '18', '74', '98', '21', '82', '11', '19', '48',
                 '53', '61', '73', '85', '96', '97', '99', '7', '44', '95', '40', '76'),
        'DA': ('16', '17', '30', '62', '93', '50', '14', '15', '39', '40', '41', '76'),
        'Graphs': ('9', '23', '29', '31', '34', '35', '45', '87',), 
        'Onto': ('1', '3', '5', '6', '13', '27', '38', '42', '67', '80', '92', '50',), 
    }
H = hnx.Hypergraph(articles)
kwargs = {
        'with_node_labels': True,
        'with_node_counts': False,
        'with_edge_counts': False,
    }

hnx.drawing.draw(
    H_collapsed,
    edges_kwargs={
        'linewidths': 2,
    },
    nodes_kwargs={
    'facecolors': colors
    },
    edge_labels_kwargs={
        'fontsize': 16,
    },
    **kwargs)
plt.show()

2. See the result:
   
   As an example, the node 76 seems to be inside the edge Onto, but it's actually not.

Expected behavior
Edges should have node inside them only when they're actually part of them.

Environment (please complete the following information):

• OS: Linux
• Python Version Py3.9

[brendapraggastis]

@Ramshell The draw function often has overlapping or planarity issues when there are a large number of nodes. There are a couple of things you can do.

1. Change the layout. Any networkx layout may be used.
2. Set return_pos=True, then tweek the returned dictionary of node positions until things line up the way you want.
3. Collapse the nodes and return the equivalence classes. This would be my choice.
   
   Equivalence classes in a dictionary:
   {'10: 17': ['10', '2', '24', '36', '4', '55', '57', '58', '60', '63', '69', '71', '75', '8', '81', '86', '91'], '11: 13': ['11', '21', '53', '61', '73', '74', '82', '85', '95', '96', '97', '98', '99'], '1: 9': ['1', '13', '27', '3', '5', '6', '67', '80', '92'], '14: 7': ['14', '15', '16', '17', '30', '62', '93'], '18: 8': ['18', '19', '33', '44', '48', '7', '84', '88'], '29: 4': ['29', '34', '35', '9'], '23: 4': ['23', '31', '45', '87'], '38: 2': ['38', '42'], '39: 2': ['39', '41'], '40: 2': ['40', '76'], '50: 1': ['50']}
   You can rename the node labels to use representatives you wish to emphasize as well.