WIP: DataCollection/StateDescriptor niceties

[bprather]

This basically adds a couple of convenience functions to the interface changes in #921, which I find useful in KHARMA. Not terribly relevant to the real content of that PR so I'm posting it separately.

Depends on #907 (at least as posted here, to prevent merge conflicts later)

Basically this adds a lot of overloads to the DataCollection::Add() function, which allow it to take strings denoting a particular MeshData to copy from, and making it default to copying from "base." I find both conventions much more readable, and I think they might be easier for newcomers to the code to understand given that allocations and Add calls are some of the first things one runs into trying to read/understand KHARMA.

It also adds a function GetPackDimension to StateDescriptor which returns the length of a theoretical pack of variables from that package given a FlagCollection. This is useful for allocating variables of the full pack length (fluxes, boundaries, etc.) before Mesh initialization.

PR Checklist

• Code passes cpplint
• New features are documented.
• Adds a test for any bugs fixed. Adds tests for new features.
• Code is formatted
• Changes are summarized in CHANGELOG.md
• CI has been triggered on Darwin for performance regression tests.
• Docs build
• (@lanl.gov employees) Update copyright on changed files

[Yurlungur]

It's going to be a little hard to read this diff until #921 is in. Can you ping reviewers once that happens?

[bprather]

Yep, definitely not worth reviewing now. I also have some changes & fixes still queued up for this, it's still evolving a little as I firm up how I'm using it in KHARMA. I'll remove WIP and ping folks when it's decent.

(May also open a different PR if git weirdness makes this one awkward, I wasn't terribly clean about branching this)