Fixed: Add section on inventory sidecar files

[ahankinson]

This PR adds a section on hash sidecar files.

Refs #29

[awoods]

Is there a canonical list of hash algorithms we can reference?

[zimeon]

See comments on #21 -- I think the answer is a bit tricky but we will need to define the list, hopefully by reference to one or more other specs. I would suggest we merge this and solve that as a separate problem

[zimeon]

Is the addition of the filename following parallel with the bagit approach? I don't object to that but it seems unnecessary

[awoods]

@zimeon : https://tools.ietf.org/html/draft-kunze-bagit-08#section-2.1.3

Additionally,

One or more linear whitespace characters (spaces or tabs) MUST separate CHECKSUM from FILENAME.

[zimeon]

Yes, I see it there and there it is necessary because the manifest-XYZ.txt file contains checksums for multiple files. However, we have a different naming convention and the checksum for just one file... so it isn't really necessary

[ahankinson]

It is also used in the output of shasum:

$>shasum -a 512 CNAME
dba3b3255390e9bda6fcc1f3cc8534a4c9fec4f90478c8e62bfaae73fab5700d524a8fb09d692c41cdfd83ec978a282082b586392325124c7cd5c21cca115599  CNAME

and is also used in software distributions as sidecars, e.g.,

http://www-eu.apache.org/dist/lucene/solr/7.4.0/

(although I'm not sure what the * in the filename means).

[zimeon]

Thanks @ahankinson - I agree the * is odd but good to see similar use in other places

[ahankinson]

I updated per #35 so re-review and approval is necessary.

[ahankinson]

@zimeon This section in the output of the shasum utility explains what the * is for:

The sums are computed as described in FIPS-180-4. When checking, the
input should be a former output of this program. The default mode is to
print a line with checksum, a character indicating type ('*' for binary,
' ' for text, '?' for portable, '^' for BITS), and name for each FILE.

So maybe we also need to decide the mode?

[zimeon]

Will port your comment above onto #39