add SoftImpute and IterativeSVD to imputation

[xinyuejohn]

PR Checklist

• This comment contains a description of changes (with reason)
• Referenced issue is linked
• If you've fixed a bug or added code that should be tested, add tests!
• Documentation in docs is updated

Description of changes

1. Add SoftImpute, IterativeSVD, MatrixFactorization, and NuclearNormMinimization from fancyimpute as imputation methods.
2. Add miceforest as imputation method.

Technical details

• pyproject.toml has been updated to add fancyimpute and miceforest dependency.
• two test data sets have been added which can be used to test imputation.
  test_impute_iris.csv only contains numerical data and test_impute_titanic.csv contains numerical data and non-numerical data.
• ehrapy/preprocessing/init.py has been updated to add import.
• tests/preprocessing/test_imputation.py has been updated to add corresponding tests.

[xinyuejohn]

Docstring needs to be updated.

[Zethson]

Looks good on a first glance. Can this also be sped up with the Intel sklearn extension?

Think that @Imipenem should review this one.

[Zethson]

@xinyuejohn you also need to add this function to the docs.

[Imipenem]

Mostly minor stuff.

[Zethson]

Please add the imputation methods to the docs (usage.rst)

[xinyuejohn]

@Zethson It seems this has no more issues.