Develop WD1 decoding

[jwaiton]

This PR introduces the functionality of processing into MULE, with single channel decoding from waveDump 1 .dat files now being possible. Addresses #11 which will be resolved when this is complete.

The data is stored in h5 files with a storage path and name provided by the user. The h5 format matches that decoded in wavedump 2.

This PR rests on top of #10, and as such includes all the new bells and whistles. Config files, integration with the mule executable.

Things still needed

• tests
• refactor the code
• remove the deprecated function

[bpalmeiro]

not real hehe

[bpalmeiro]

request

[jwaiton]

Sorry for bloating PRs, but in this PR I'm including a smarter way to process data that is memory efficient. We now have a writer object that can writes individual waveforms to the h5 file.

I need to add forwards compatible functions for reading datasets lazily now, and make them group and dataset agnostic. This is attached to #31, and will be done in a different PR.

EDIT: To avoid this getting out of hand, I'll add tests for the readers and writers in a separate PR, or perhaps I'll extract them into a separate PR entirely (the smarter but possibly more complicated move git-wise).

EDIT 2 (29/03): The readers and writers are also a bit stupid, and waste a lot of time with IO operations (reopening datasets that should rest open at all times). I tried to think of a smart solution to keep them open but it gets complicated (or I'm tired, probably both).

[jwaiton]

Test coverage now near 100%, although works on a 'whole function' basis, due to the reader, writers, and the decoder having limited compartmentalisation. Improved functionality would be nice, but currently it works as expected. I'd say its ready for a 'light' review, with the knowledge that almost every issue found will become technical debt :)

[jwaiton]

I've split this PR up into two separate ones (#40 & #41), to help reviews flesh out more easily.