Figure out consistent strategy for error handling in resource detection

[psx95]

There needs to be a consistent strategy to follow error handling within the resource detection code in case the code reaches unexpected conditions.

As of now, all exceptions & unexpected conditions are either swallowed or simply ignored.

[aabmass]

From @dashpole: sometimes we log errors + swallow and sometimes throw exceptions. Would be nice to come up with a strategy here before working on it.

[damemi]

Moving to p3