Missing value imputation

[chris-1187]

User Story

1. As a RTDIP user
2. I want my time series data to be consistent in the pattern I expect and not incomplete
3. So that my BI functions and planned doings on this data later is enhanced by the provision of a curve of continuous data points

Additional context

• This component should work with time series data with a fixed pattern
• It should use imputeFD or MICE, which are common ML algorithms / methods (based on which is more relevant to the usecase)
• This can be done by integrating Apache SystemDS through its python bindings, which is a perfect fit because it runs on top of Spark and is optimized for single node processing

Acceptance Criteria

• A new component is implemented
• Component input
  • live data (adjusted and filled with missing data points as NaN value)
  • the name of column with the value data that should be imputed
• Component output
  • the same dataframe of live data with imputed values for continuous data points

Definition of Done

• Test cases have been created and are runnin successfully
• Documentation for the new component was added
• Github Actions are running without errors