- Supports float64 computations (Can become useful in some cases)
- Use approximate model for SKT also (future finite difference)
- Add a parameter to set the initial prior std (And correct for Brownian KOFT)
- Small docs improvements