Riskfolio-lib

=============

- Add two new convex risk measures: EVaR Range and RLVaR Range, to the Portfolio object.
- Add three new risk measures: VaR Range, EVaR Range and RLVaR Range, to the HCPortfolio object.
- Add the generalization of risk parity for variance through inequality constraints on the risk contributions of assets to the Portfolio object.
- Add the generalization of factor risk parity for variance through inequality constraints on the risk contributions of risk factors to the Portfolio object.
- Add a function to calculate the Brinson Performance Attribution per class and aggregate.
- Add a plot function to show the Brinson Performance Attribution.
- Add functions to calculate the VaR Range, EVaR Range and RLVaR Range.
- Update plot functions to consider EVaR Range and RLVaR Range.
- Update duplication, elimination and summation matrices functions to consider or not the diagonal of the symmetric matrix.

=============

- Add new functions to calculate the number of effective assets (NEA) and the average centrality of the portfolio.
- Add the possibility to use neighborhood and cluster network constraints at the same time.
- Fixed some bugs in HRP and HERC when we add constraints.
- Fixed a bug in the duplication_summation_matrix.
- Fixed tight layout in plot functions that uses multiple axes.

=============

- Improvement in calculation speed of duplication_matrix, duplication_elimination_matrix and duplication_summation_matrix functions using a vectorized formula.
- Fixed formulation of risk parity with risk factors model that produced incorrect results when using the MOSEK solver.
- Fixed some bugs in PlotFunctions module.
- Fixed some bugs in HCPortfolio related to custom_mu vector and use of Kurtosis and Semi Kurtosis as risk measures.
- Standardized the way additional parameters to estimate mean vector and covariance matrix are entered.

=============

- Implements standarized silhouette score to determine the optimal number of clusters.
- Fix plot_clusters function to plot clusters and heatmap in same order of codependence matrix. Originally it plots the codependece matrix with axis x inverted.

=============

- Implements risk parity optimization based on explicit risk factors and principal components.
- Implements new formulations of Gini Mean Difference, Tail Gini, Range, CVaR Range and Tail Gini Range that improves speed compared to formulations based on the owa portfolio model.
- Improves the calculation of elliptical uncertainty sets for worst case optimization.
- Add new functions that allow us to calculate the risk contribution per explicit risk factors and principal components.
- Add new functions that allow us to plot the risk contribution per explicit risk factors and principal components.

=============

- Implements new kind of constraints that incorporates the information from networks like the Minimum Spanning Tree and Maximally Filtered Graph into the portfolio optimization models: return-risk portfolio, owa portfolio and worst case portfolio.
- Implements new kind of constraints that incorporates the information from dendrograms into the portfolio optimization models: return-risk portfolio, owa portfolio and worst case portfolio.
- Improves the speed of several functions using the c++ linear algebra library Eigen and c++ eigenvalues library Spectra.
- Add new functions that allow us to plot the relationship between graphs and asset allocation.
- Add new functions that allow us to create constraints based on graphs information.
- Add a new example about applications of networks and dendrograms constraints in portfolio optimization problems.
- Fixed some errors related to HCPortfolio with constraints.
- Fixed some errors in some plots.