Use tabulated profiles for `massDistributionSphericalHeated` if using tidal heating

[abensonca]

If the profile to be heated is of the massDistributionSphericalTabulated class, then we know that tidal heating just adds one more dimensionless parameter, $Q r_\mathrm{s}^2/(\mathrm{G} \rho_\mathrm{s} r_\mathrm{s}^3/r_\mathrm{s}$). We could automatically expand the tuple of tabulation parameters (and wrap the factory method that builds profiles during tabulation) to include this extra heating parameter. This should allow for automatic tabulation of heated, tabulated profiles.

If coupled with converting the NFW profile to a tabulated case then this might lead to significant speed-up for dark matter only models.