Unfortunately today there isn't an easy way to this. You can do it a single device at a time, or globally. In my mind (and past practice), I say do the global reset. (1) You will automatically (attempt to) pick up more from the unknown table. (2) You will ensure that all devices that have changed will automatically be re-configured, as opposed to a grouping or other mechanism where some might fall through the cracks. (3) You validate that all of the devices you expected to be able to access via SNMP are in fact accessible (still?).
In other words, I agree it perhaps is not the ideal answer for you, but from my perspective, I see enough benefit in this approach to go for the larger, 'overkill' answer. I guess the only caveat would be in an environment where I have devices (like UPS's) that are highly intolerant to multiple SNMP connection attempts.