Add ttest_equal_var argument to control t-test variance assumption

[tompollard]

This pull request introduces a new ttest_equal_var argument to the TableOne class, allowing users to specify whether equal variances should be assumed when performing t-tests (see: #193).

By default ttest_equal_var=False, which performs Welch’s t-test, a test that does not assume equal variance between groups.

df = pd.DataFrame({
    'group': ['A', 'A', 'A', 'B', 'B', 'B'],
    'x': [1.0, 2.0, 3.0, 20.0, 22.0, 24.0]
})

ttest_equal_var=False

t1 = TableOne(df, columns=['x'], groupby='group', pval=True, ttest_equal_var=False, pval_digits=5)
pval_welch = t1.tableone[('Grouped by group', 'P-Value')].iloc[1]

# pval_welch == "0.00065"

ttest_equal_var=True

t2 = TableOne(df, columns=['x'], groupby='group', pval=True, ttest_equal_var=True, pval_digits=5)
pval = t2.tableone[('Grouped by group', 'P-Value')].iloc[1]

# pval == "0.00010"