Sort data_files alphabetically when using sc_indicate and combine_to_ind

When using `combine_to_ind(ind_file, ...) ` (which wraps `sc_indicate` as `sc_indicate(ind_file = ind_file, data_file = c(...))` to match the pattern needed for combiners in task tables), we often get hash tables where the files are in different orders for different builds. This becomes clear when the hash table files are checked into git, a la [this](https://github.com/USGS-R/lake-temperature-model-prep/pull/247/files#diff-2bdaddf796bedeb750788ab0c210f414cc8ca644c2d4b599e1a9e20d7a864cfa)

This makes it harder to tell if the files have changed (hash change) or just the order has changed. The changes would stand out more if the order was consistent.

Start by locating the implementations of combine_to_ind and sc_indicate and trace how data_file values become hash table files. Sort the relevant data_files alphabetically for this path, then verify that repeated builds produce the same file order and hash contents.

Written by the indexing model from the issue text.