Maintaining the association between two variables (one of them multiply comma-delimited)

Magic Banana's first solution (Wed, 02/23/2022 - 18:56) does exactly what I want.

I actually believe the second solution is better: it complements the existing table instead of creating a new incomplete one.

To get the same format as that of my earlier command but using the completed table:
$ cut -f 1,8 completed_file | awk -F '[,\t]' '$2 { for (i = 2; i <= NF; ++i) print $1, $i }'
That should actually be in a separate executable script, maybe named unflatten-pairs, expecting a field number (8 in the command line above) and optionally files, that you would save in a directory listed in your PATH variable:
#!/bin/sh
if [ "$1" -gt 1 ]
then
field_pos=$1
shift
cut -f 1,$field_pos "$@" | awk -F '[,\t]' '$2 { for (i = 2; i <= NF; ++i) print $1, $i }'
exit
fi
printf "Usage: $0 field_pos [file]...
Every output line is a value in the first input column associated with
every comma-separated value in the field_pos-th input field.
"
In this way, to get Hairball-file.txt's URLs (for instance) in the other format, you would simply write:
unflatten-pairs 6 Hairball-file.txt
That is what people writing minimally-reusable code would do at least. I wrote "minimally-reusable" because the script could be made more general, not hard-coding the tabulation and the comma as the separator, and performing a Cartesian product of an arbitrary number of columns.

What I've been doing is placing zeros between back-to-back tabs to mark empty fields with sed

That makes further processing harder. If a field is empty, why not keeping it empty?

That requires that we first list each UID followed by the first email, then another UID followed by the second email associated with that UID, etc. and then separating the domain from each email to obtain a two-column file of UID's and domains (from Col$7), just like the UID's and domains from the many URL's in Col$6 of Hairball-file.txt.

If it is "just like the UID's and domains from the many URL's" (it does not look like, from what you wrote before), then the solution simpler: there is nothing to remove after the domain. To complement Hairball-file.txt (or Hairball-file.txt already completed with the distinct domains in the URLs), you can write:
$ cut -f 7 Hairball-file.txt | awk -F , 'gsub(/[^,]*@/, "") { for (i = 1; i <= NF; ++i) if (!($i in a)) { a[$i]; s = s "," $i }; delete(a) } { print substr(s, 2); s = "" }' | paste Hairball-file.txt -
If you then want the other format, you can post-process the above command line with unflatten-pairs (see the beginning of this post).