update loc_table to utilize top-down pull approach

[d-hergott]

The current loc_table in ramptools uses a branching approach that starts at the health facility and then pulls in the parents to those health facilities. This results in some subcounties, which don't have health facilities, not included in this version of loc_table. This becomes problematic when trying to link with shapefiles or calculate overall statistics for a district (population, subcounty count, etc.)

Therefore, will replace with a version created from a "top-down" approach, in which the children of each region is pulled. Austin pulled this on 08MAY2024.

[d-hergott]

I merged if with the bottom_up approach (also pulled in May) to get a combined file. Script for this is in my Scratch folder in Box. loc_table_merge_May2024_version

This combines the bottom up loc_table pulled on 240508_location_hierarchy with the top-down location table pulled 240524_location_hierarchy.

Updated loc_table has the information below.

> loc_table[,.N,level]
   level     N
   <int> <int>
1:     1     1
2:     2    15
3:     3   146
4:     4  2312
5:     5  8504

[d-hergott]

We've discussed splitting the loc_table into lt and hf (so separating the health facilities out). However, there's still an issue with the levels changing.

One option might be to save the levels as objects within ramptools, and then call those in the scripts?

For example, district_level= 5 and then in the code, would have loc_table[level==district_level]