Rewrite GuppiRaw() reader to not raise EndOfFileError and use Context Management about blimpy

[telegraphic]

Changed title and added enhancement label. Relevant reading:
https://jeffknupp.com/blog/2016/03/07/python-with-context-managers/

[jeenriquez]

Greg, Danny, could you check on this? or can we close it?

[telegraphic]

This basically requires a rewrite of the GuppiRaw reader, so can't close it yet...

[telegraphic]

Ok, I think it would be nice if we could run:

with GuppiRaw(filename) as raw:
    for block in raw:
        print(block.header)
        print(block.data.shape)

I think this may need a yield function? Need to make sure that data are only loaded when needed, not at init