Thank you Steve, Joyce and Patty for your comments and thoughts.
SEMs reveal the micropores are only in the weld areas and it is presumed by all including the vendor that the plating did not have the throwing power to coat and seal the floor of the microvoids. Depending on processing, there is a possibility that the entrapped moisture was a remnant of processing and arrived in the sealed from the factory boxes.
These points remain...
History has it that military products liked gold because it offered indefinite shelf life.
Custom packagers of low volume products (e.g., military) need to carry some inventory to be responsive.
Ceramic is most often used where hermeticity and heat resistance are desired. I agree that 1.5 microns is perhaps a bit too much for soldering but it is a good number of gold wire bonding and that is the way that the parts are apparently coming in. (In former times when I work for a Mil Aerospace company, we pre-tinned gold leads before assembly to mitigate the former concern.)
While the discoloration is not likely to cause a problem, it is indicative of a metal finishing process that is not in control in my opinion. (I am now sharing some of my thoughts)
I have gold plated circuit boards, with much less gold, that have been hanging on the wall in an uncontrolled environment for more than 20 years and they do not exhibit this condition in any remote fashion.
Were I the vendor, I would acknowledge the simple and obvious facts and remedy the situation both in the process and with the customer. But I guess I am showing my age... ;-)
Thanks again to all and very best regards,
Joe