This would seem to be a question ripe for an answer--good for innumerable PhD theses, papers, reports...except for some reason the data is not available. I wonder why.
--Boris
On this particular question, I've been quoting the same Web site/page for several years. It has a few references.
Gases: Man versus the Volcanoes
It covers CO2, SO2, HCl, HF, and HBr. About the only other one of possible significance might be H2S (and that from steam vents around volcanoes, not an "actual" magmatic gas).
As you say, this is a research topic that generates quite a bit of data; but most researchers focus on one or a few volcanoes rather than compiling global inventories. Part of the problem is that volcanic emissions are highly variable -- you would have to do a lot of adding over a lot of years to get improved estimates.
I've noted previously that the author of some of these papers -- Gerlach -- seems to be one of the few that does "global" work regularly. According to what I kind find about him on the Web (not much) he probably is still working at the Cascades Volcano Observatory (USGS).