Fixed imputation order handling in ImputationKernel

[nmehran]

This pull request addresses a bug in the ImputationKernel class related to the imputation_order parameter. The issues fixed include:

• Typo Correction: "decending" was corrected to "descending".
• Sorting Configuration: Updated sorting configuration for ascending order to use ascending=True.

Steps to Reproduce

The following code can be used to test and verify the changes:

import pandas as pd
import numpy as np
import miceforest as mf
from sklearn.datasets import load_iris

# Load the Iris dataset
iris_data = load_iris()
iris_df = pd.DataFrame(iris_data.data, columns=iris_data.feature_names)

# Introduce missing values
np.random.seed(42)
missing_indices = np.random.choice(iris_df.index, size=10, replace=False)
iris_df.loc[missing_indices, 'sepal length (cm)'] = np.nan
iris_df.loc[np.random.choice(iris_df.index, size=8, replace=False), 'sepal width (cm)'] = np.nan
iris_df.loc[np.random.choice(iris_df.index, size=4, replace=False), 'petal length (cm)'] = np.nan
iris_df.loc[np.random.choice(iris_df.index, size=2, replace=False), 'petal width (cm)'] = np.nan

# Initialize ImputationKernel with 'ascending' order
kernel_ascending = mf.ImputationKernel(iris_df, imputation_order='ascending')

# Initialize ImputationKernel with 'descending' order
kernel_descending = mf.ImputationKernel(iris_df, imputation_order='descending')

# Compare imputation orders
ascending_order = kernel_ascending.imputation_order
descending_order = kernel_descending.imputation_order

# Print equality comparison result
print("Are the imputation orders equal?")
print(ascending_order == descending_order)

Related Issue

This PR resolves issue #90.

[AnotherSamWilson]

Nice catch

[AnotherSamWilson]

This commit was merged as a part of this PR.