DatasetPage optimization cleanup, another pass #9762
Description
In #9725/#9736 we addressed some performance issues that were serious enough to be considered release showstoppers. I had to stop there because we needed to release, but there are still lots of improvements to be made/lots of waste to cut in that page, and I'd love to address it soon.
Specifically, there are numerous places in the backing bean where Service.find(dbId) is executed in order to look up datafiles and dataverses, instead of using the already available entities (that we now go to the trouble of initializing in a more efficient manner early on). Unfortunately, a few more were added in the last release. Unlike some cases addressed in #9736, these remaining lookups are performed only on the current page-worth of files (not on every file in the dataset/version), so the consequences are not as catastrophic. But it still adds up.