Fixing handling of pivot

[Edwardvaneechoud]

This pull request refactors the location of the calculate_fuzzy_match_schema and pre_calculate_pivot_schema functions, updates the pivot schema logic, and improves test coverage for pivot-related functionality. The main changes involve moving schema-related logic to a more appropriate module, refining how pivot output columns are named, and adding new tests to ensure correct schema calculation.

Refactoring and code organization:

• Moved calculate_fuzzy_match_schema and pre_calculate_pivot_schema from flow_data_engine/fuzzy_matching/settings_validator.py to a new schema_callbacks.py module for better separation of concerns and maintainability.

Pivot schema logic improvements:

• Updated the logic in pre_calculate_pivot_schema to generate simpler column names (e.g., '0', '1', etc.) when only one aggregation is used, instead of appending the aggregation name (e.g., '0_sum').
• Adjusted test assertions to match the new column naming convention for pivot outputs.

Testing enhancements:

• Added new tests test_pivot_schema_callback and test_schema_callback_in_graph to verify correct schema calculation for pivot operations and ensure the new schema logic is functioning as intended.
• Updated imports in the test suite to reflect the new location of schema callback functions.