Fix Single He star inlist layering

[sgossage]

The inlist layering in single star grids was modified in the past to avoid overwriting any provided star1_controls_user. However, that patch took an approach that was incompatible with subsequent star formation scenario inlist layering. In construct_static_inlist,

        for step in range(number_of_star1_formation_steps):
            star1_formation['step{0}'.format(step)] = {}
            star1_formation['step{0}'.format(step)]['inlist_file'] = os.path.join(working_directory, 'star1', 'inlist_step{0}'.format(step))
            for k, v in star1_formation_dictionary.items():
                star1_formation['step{0}'.format(step)][k] = v[step] if type(v) == list else v

will look for star formation settings in mesa_inlist['star1_job_*'] and mesa_inlist['star1_controls_*']. The previous fix added star1_controls_special as a separate, single-item entry rather than adding the special inlist to the existing star1_controls_user sequence. As a result, the special inlist was not layered consistently with the star-formation scenario inlists, and the resulting sequence could be incomplete.

This PR instead keeps the existing star1_job_user and star1_controls_user dictionary items and removes the separate star1_controls_special entry. It prevents the original overwrite by modifying the line that caused the issue:

    if system_type == "HMS-HMS" and mesa_inlists['single_star_grid']:
        ...
        mesa_inlists['star1_controls_user'] = special_single_star_user_inlist

changing this to

        if system_type == "HMS-HMS" and mesa_inlists['single_star_grid']:
            ...
            if 'star1_controls_user' not in mesa_inlists or mesa_inlists['star1_controls_user'] is None:
                mesa_inlists['star1_controls_user'] = []

            mesa_inlists['star1_controls_user'].append(special_single_star_user_inlist)

This preserves any existing star-formation inlists while adding the special single-star control inlist, following the same layering approach already used for the CO-He_star scenario.

[maxbriel]

One question, but otherwise looks good

[maxbriel]

Are you sure that this is a list already? I thought that it was initialised as a string (with the filepath)

[sgossage]

In the case of single HMS I'm sure, but actually we need to add something more to the single HeMS case.

In general, mesa_inlists is first defined in main() after parsing the .ini file.

Later in find_inlist_from_scenario, this dictionary gets modified in various ways. Focusing on the single stars though:

• In the single HMS case, that if statement above is the first check since reading the ini file as to whether or not mesa_inlists['star1_controls_user] exists or not (or if it is set to None). If it does not exist, we need to create it, because we need it to hold at least special_single_star_user_inlist. So, if isn't in the .ini file, it is initialized as an empty list here, then star1_controls_user is always appended to that list, or whatever pre-existing list there may have been.
• In the single HeMS case, there's a line, actually two, that force star1_controls_user to a specific set of inlists:

348            mesa_inlists['star1_controls_user'] = single_star_scenario
349            mesa_inlists['star1_job_user'] = single_star_scenario

so that will overwrite anything that was set in .ini in the case of the single He star grid.

I've now added a similar if statement in the single HeMS case now that checks if these dict items exist already, and if so, append single_star_scenario, and subsequently, append special_single_star_user_linlist as well.