Add option to fit sed with G23 only

[Chamani8]

The full pull request includes two commits.

The first commit adds an argument to dust_extinguished_sed() function in model_data.py:212, to take in a keyword that allows the sed to only be fit with G23().

dust_extinguished_sed() is called by lnlike() in fit_model.py:88, which in turn is called by lnprob() in fit_model.py:151
both lnlike() and lnprob() are updated to take in the new keyword "g23" as an optional argument. These were made optional arguments as not to break anything that continues to use these functions without the new argument. The default keyword is set to "all", which executes the original sed fitting in dust_extinguished_sed(), rather than G23 only. The original sed fitting, done using the _curve_F99_method() can be called by excluding the new argument, when calling lnlike(), lnprob(), and/or dust_extinguished_sed().

The second commit adds a warning if mismatched band information is provided, between the model and reddened star data.

[karllark]

Looks good. A couple of changes needed.

[karllark]

Instead of having this as the "elif", should it be "else"? In other words, what happens if someone passes fit_range = "all_butItypeitwrong"? Or an "else" statement needs to be added to handle the case with fit_range is not "all" or "g23".

[karllark]

Codestyle errors need to be fixed. Can run python -I -m pip install flake8 in the base repository directory to find the errors. I use black to format code so that I don't have to manually fix these issues.

[karllark]

doc test error is fixed in PR #123

[Chamani8]

Following updates have been made:
modeldata.py:212 added docstring for the fit_range argument.
modeldata.py:240,246 testing fit_range with all lowercase.
modeldata.py:283 else clause raises ValueError letting users know the correct options for fit_range argument.
both modeldata.py and fit_models.py have been run with black to fix codestyle errors.

[karllark]

LGTM