NA POIs in refactor output

This may be related to Issue #114 (closed). In the refactor output, there are mapped_POIs with an NA identifier.

For example in refactor_01.gpkg I see:

> gpkg = "/Volumes/Transcend/ngen/CONUS-hydrofabric/refactor/refactor_01.gpkg"

>   read_sf(gpkg, 'mapped_POIs') %>% 
+     filter(is.na(identifier))

Simple feature collection with 586 features and 22 fields
Geometry type: POINT
Dimension:     XY
Bounding box:  xmin: 1848139 ymin: 2218321 xmax: 2255727 ymax: 2988610
Projected CRS: NAD83 / Conus Albers
# A tibble: 586 × 23
         ID  toID set   ident…¹ COMID Type_…² Type_…³ Type_TE Type_…⁴ Type_…⁵
 *    <dbl> <dbl> <chr>   <int> <chr> <chr>   <chr>   <chr>   <chr>   <chr>  
 1 10002771    NA 1000…      NA NA    NA      NA      NA      NA      NA     
 2 10003160    NA 1000…      NA NA    NA      NA      NA      NA      NA     
 3 10003865    NA 1000…      NA NA    NA      NA      NA      NA      NA     
 4 10003899    NA 1000…      NA NA    NA      NA      NA      NA      NA     
 5 10004013    NA 1000…      NA NA    NA      NA      NA      NA      NA     
 6 10007677    NA 1000…      NA NA    NA      NA      NA      NA      NA     
 7 10009039    NA 1000…      NA NA    NA      NA      NA      NA      NA     
 8 10009236    NA 1000…      NA NA    NA      NA      NA      NA      NA     
 9 10009238    NA 1000…      NA NA    NA      NA      NA      NA      NA     
10 10014509    NA 1001…      NA NA    NA      NA      NA      NA      NA     
# … with 576 more rows, 13 more variables: Type_WBOut <chr>,
#   Type_Conf <chr>, Type_Term <chr>, Type_Elev <chr>, Type_Travel <chr>,
#   nexus <chr>, snapped <lgl>, TotDASqKM <dbl>, DnHydroseq <int>,
#   type <chr>, member_COMID <chr>, Type_Con <dbl>, geom <POINT [m]>, and
#   abbreviated variable names ¹​identifier, ²​Type_HUC12, ³​Type_Gages,
#   ⁴​Type_NID, ⁵​Type_WBIn
# ℹ Use `print(n = ...)` to see more rows, and `colnames()` to see all variable names