[Roadmap] Malisani transformation — regularization of DAE optimal control problems

## Objective

Build a system to construct the Paul Malisani transformation for regularization of DAE optimal control problems.

## Context

The [BVP-DAE](https://github.com/control-toolbox/BVP-DAE) repository is the starting point. The goal is to provide a nice, usable system to apply this regularization approach within the control-toolbox ecosystem.

## Why

The Malisani transformation is a powerful regularization technique for optimal control problems with DAE constraints. Having a dedicated tool will make this approach accessible and integrate it with the rest of the ecosystem (CTFlows, CTIndirect, CTSolvers).

## References

- [BVP-DAE repository](https://github.com/control-toolbox/BVP-DAE)
- [Malisani et al., HAL 2020](https://hal.science/hal-04191269)
- [Roadmap comment](https://github.com/orgs/control-toolbox/discussions/68#discussioncomment-17584843)

Start by studying the BVP-DAE repository and the referenced Malisani et al. paper to understand the proposed transformation and existing proof of concept. Review the roadmap discussion for project expectations. Done means providing a usable regularization system that integrates with CTFlows, CTIndirect, and CTSolvers.

Written by the indexing model from the issue text.