feat: Baseline Data Cleaning

[zogomii]

Is your feature request related to a problem?

Subtask of #710

Desired solution

Create method Baseline._clean(table: Table, target_column: str)->TabularDataset for baseline data cleaning

1. Remove columns with high idness or stability (either above 90%), excluding the target column
2. Remove columns with high missing value ratio (above 60%)
3. Impute all remaining columns with missing values using highest (absolute) correlating column
4. One hot encode all non-numerical columns with less than 20 different values, remove all other non-numerical columns
5. Remove outliers
6. Normalise columns with values greater than 100

Possible alternatives (optional)

No response

Screenshots (optional)

No response

Additional Context (optional)

No response