Revised core ULDM halo class

[AlexLaroche7]

• Removed TNFW + ULDM option, now only CNFW + ULDM, also removed the corresponding test file/functions
• Keyword argument specifying component eliminated from preset model
• Changed optimization algorithm to SLSQP so that bounds can be enforced. This avoids negative values for q and beta and improves the mass range over which the function works correctly.
• Added mass and density tests for the ULDM halo class. Mass is accurate to 1% and density is accurate to 2%.
• Added documentation to optimization functions, and explicitly wrote out units in Schive functions.

[AlexLaroche7]

Added the following:

• ULDM example notebook
• 'log_mlow' modification to ULDM realization so that no halos are created below input 'log_mlow'
• 'scale_nfw' boolean parameter specifying whether or not to adjust alpha_Rs (for mass accuracy)

Even when 'scale_nfw' = False, density profiles are not exactly equivalent because the cored radius of the CNFW profile changes the profile slightly. You can see how different the profiles are at large radii in the example notebook. Perhaps this is something we will need to correct later on.