Initial support for sparse matrix representation of H-polytopes

[ChinmayRout9040895625]

This PR introduces the initial infrastructure required to support sparse matrix representations of H-polytopes in Rvolesti. While the volesti C++ backend already supports sparse matrices, the R interface currently lacks a pathway for handling sparse inputs efficiently.
The changes focus on enabling R-to-C++ transfer of sparse matrices using RcppEigen and preparing the package for future integration with the sparse polytope backend.

Changes

1- Updated S4 class definitions to allow sparse matrix storage where appropriate
2- Added dependency on the Matrix package for sparse matrix support
3- Linked RcppEigen to enable automatic conversion of R sparse matrices (e.g., dgCMatrix) to Eigen sparse matrices
4- Implemented a minimal C++ wrapper (create_sparse_hpolytope) to verify successful transfer of sparse matrices from R to C++
5- Added basic tests for sparse input handling

Notes

1- This PR does not yet connect to the full volesti sparse polytope implementation; it establishes the foundational infrastructure required for that integration.
2- Existing functionality for dense H-polytopes remains unchanged.
3- Backward compatibility is preserved.

Supporting sparse matrices is important for high-dimensional polytopes where the constraint matrix is largely sparse. This can significantly improve memory usage and computational efficiency for large-scale problems.

Future work will focus on:

1- Constructing volesti sparse polytope objects from the transferred matrices
2- Extending sampling and volume estimation functions to support sparse inputs
3- Performance validation against dense representations

Related Issue

Closes/Relates to: #29 (Support sparse matrices to represent H-polytopes)

Thank You!