feat(Prophet-coeffs): DS-2469 store prophet coeffs 2

[Eliel-Avilez]

Please refer to this Robyn 2 PR which is part 1 of this development.

• Quick Recap of the above mentioned PR: A new output file for Robyn was added, called prophet_regressor_coefficients.csv which stores the Prophet regressors and the coefficients of the factor variables, example:

	regressor	coefficient
1	Marathons_0	-0.014
2	Marathons_1	0.006
1	RandomVariable_X	-0.003
2	RandomVariable_Y	0.006

Changes

• Added the columns factor_var, and level to the output file. We could encounter scenarios where some factor name is contained within other factor name. This two new columns will give us clarity of the name and the values of the variable instead, so we assume as little as possible, example:

	factor_var	level	regressor	coefficient
1	events	black_friday	events_black_friday	-0.014
2	events	spring_break	events_spring_break	0
1	events_legacy	independence_day	events_legacy_independence_day	0
2	events_legacy	memorial_day	events_legacy_memorial_day	0.006

• Changed the way we get the present regressors that Prophet estimated when adding the reference levels. (inputs.R/add_reference_levels_to_prophet_coefficients()).

  • Before: Concatenating factor var names and levels to form present regressors names for each factor.
  • After: We get the present regressors for each factor by checking the existence of the general present regressors list in the list of OHE columns formed for a specific factor.

  With this, we avoid the possibility of wrongly assigned levels for each factor if factor names are alike, e.g. events: black_friday, spring_break, independence_day, memorial_day.

[GG-TechOpsSuper]

✅ Snyk checks have passed. No issues have been found so far.

Status	Scan Engine	Critical	High	Medium	Low	Total (0)
✅	Open Source Security	0	0	0	0	0 issues
✅	Licenses	0	0	0	0	0 issues

💻 Catch issues earlier using the plugins for VS Code, JetBrains IDEs, Visual Studio, and Eclipse.