Use default archive path in notebooks

[sbenthall]

The way things are supposed to work now is this:

• A configuration file fixes the archive directory relative to the installation path of BigBang
• When collect-mail is ran, it stores the archive data in this directory
• When an Archive object is loaded, it by default goes to this directory, so that path is determined by a short name.

The Examples notebooks were written before this was implemented, and so are all over the map with respect to how they choose the Archive path.

Moreover, now that we more explicitly support installation of BigBang via pip and not via git cloning, it's not clear to me that this way of configuring BigBang is stable.

We should try to figure out a more pain-free way of doing this. Checking in with @MridulS about what he did to get the dashboard data in one place would be a good idea.

[MridulS]

I need to rethink about this bit but we should definitely be revisiting the archive/config paths. That was a serious pain point in trying to package the dashboard. I'll add it to my TODO 😅

[laurenmarietta]

A related thing that I've just ran into is that collect-mail (in populate_provenance()) requires the --archives option to point to a path within a git repo. I'm currently trying to download IETF etc. archives to an external hard drive (and I imagine, given the size of the mailing list data, that others might try to do the same) - so while address this path question it would perhaps be useful to include in an option for a non-git path.