Given that, I think it would be a good thing to generalize the -d switch (and maybe the --maintenance-db switch too).
Just a couple of quick comments:
Some of those tools user --dbname as a long option.
You're right. I checked and each and every tool that allow the -d switch allows the --dbname. So, of course, if -d is implemented for all Postgres client, --dbname should be allowed too.
Most of those tools also use the connection environment variables used by libpq: PGDATABASE
Pgbench is documented [1]: pgbench [option...] [dbname]
Maybe I wasn't clear enough. My point was using a connection string is not documented. Here is PgBench documentation about dbname:
> where dbname is the name of the already-created database to test in. (You may also need -h, -p, and/or -U options to specify how to connect to the database server.)