corrected labels and pos args in conf_mat

[lshpaner]

Description:

This pull request resolves the issue of misaligned "Pos" and "Neg" labels in the confusion matrix printout. The previous implementation incorrectly labeled the confusion matrix elements, leading to confusion in interpreting the results. This fix ensures that the labels correctly align with the corresponding values in the confusion matrix.

Related Issue: Closes Issue #32

Previously, the constructor had the following labels:

self.labels = ["tp", "fn", "fp", "tn"]

Changes Made:

• Updated the self.labels list to correctly reflect the confusion matrix elements

self.labels = ["tn", "fp", "fn", "tp"]

• Modified the _confusion_matrix_print function to properly align the "Pos" and "Neg" labels with the corresponding values in the confusion matrix:

def _confusion_matrix_print(conf_matrix, labels):
    max_length = max(len(str(conf_matrix.max())), 2)
    border = "-" * 80
    print(border)
    print(f"{'':>10}Predicted:")
    print(f"{'':>12}{'Pos':>{max_length+1}}{'Neg':>{max_length+3}}")
    print(border)
    print(
        f"Actual: Pos {conf_matrix[1,1]:>{max_length}} ({labels[3]})  {conf_matrix[1,0]:>{max_length}} ({labels[2]})"
    )
    print(
        f"{'':>8}Neg {conf_matrix[0,1]:>{max_length}} ({labels[1]})  {conf_matrix[0,0]:>{max_length}} ({labels[0]})"
    )
    print(border)

If you try and rerun the return_metrics on the example AIDS Clinical Trials Notebook, you will see that the issue is fixed, and the correct labels will be provided:

Confusion matrix on set provided: 
--------------------------------------------------------------------------------
          Predicted:
             Pos   Neg
--------------------------------------------------------------------------------
Actual: Pos  64 (tp)   40 (fn)
        Neg  20 (fp)  304 (tn)

[elemets]

Just tested on my machine on the breast cancer data. The confusion matrices and the classification report now match so this fix works. Merging with main.