Fix incorrect edges in adjacency_matrix_to_graph

[rahulbshrestha]

This PR fixes the issue of incorrect edges in adjacency_matrix_to_graph when the adjacency matrix is of type np.matrix by converting the matrix into a numpy array.

Related issue: #887

[rahulbshrestha]

(Copying from another comment) I was a bit confused about the direction of the edges. If the adjacency matrix is

np.matrix(
    [[0, 0, 1, 0, 0],
     [0, 0, 1, 0, 0],
     [0, 0, 0, 1, 1],
     [0, 0, 0, 0, 0],
     [0, 0, 0, 0, 0]]
)

the graph is currently,

isn't it supposed to be:

because the typical interpretation of adjacency matrices is rows represent source nodes and columns represent destination nodes. Is this a bug or a conscious decision?

This comes from the for-loop,

    for to, from_, coef in zip(dirs[0], dirs[1], adjacency_matrix[idx]):
        d.edge(names[from_], names[to], label=str(coef))

[rahulbshrestha]

@amit-sharma If the design choice for the direction of the edges (adjacency matrix -> graph) is intentional and supposed to be the way it is, then the PR is ready to be merged!

[amit-sharma]

thanks @rahulbshrestha . We can take up the direction of adjacency matrix in a separate thread. Merging as is.