Fix NSBH disk mass fit

[atoivonen13]

use BH spin instead of chi_eff for NSBH disk mass fit

[atoivonen13]

We need to decide whether to modify the current function chieff2risco, or just pass the spin of the BH to that function in this case (as I have done in the PR so far)

[tsunhopang]

as @atoivonen13 suggested, we would indeed need to adjust the usage of the risco, since in the remnant_disk_mass_fitting we are using it as the normalized one, but in dynamic_mass_fitting we are using it as the risco

[atoivonen13]

@tsunhopang In the end, the only difference comes down to passing chi_eff or chi_bh. Normally, I would simply change the function to take an argument of chi instead of chi_eff and leave a function description saying passing chi_eff gets you risco while chi_bh gets you the normalized spin. However, we would also then probably want to rename the function chieff2risco to chi2risco, which could be an issue for any other uses of this function. Otherwise we can simply leave a note in the description saying passing chi_bh gives you the normalized risco. Duplicating the function seems unnecessary as the code is the same.

[tsunhopang]

@atoivonen13 could u take a look and see why the two unit tests are failing? After these are fixed I will have this PR merged

[tsunhopang]

Lgtm