[jmcvey3]

Solution for issue #470.

So far I've implemented code that allows this new data packet to be skipped. It will read the bottom track header variables, but not the bottom track velocity or distance variables. I will most likely need to reverse engineer to uncover these variables.

[jmcvey3]

It looks like Nortek reordered how variables are stored in the new bottom track data version, but I have the bottom track velocity, distance and FOM getting pulled from the file.
I also cleaned up the indexing tool in nortek2_lib.py, so it should be more robust to unknown data packets in the future

[jmcvey3]

Corrected the ensemble counter for the index file after discovering previous fixes had "over-corrected" the problem. This allows for multiple of the same ping type to occur even if they have the same stored ensemble number (usually "1").

Also discovered Dolfyn was averaging some of the 1D variables that are saved for each profile type, in this case the 5th-beam-specific variables (1D variables with the "_b5" tag). That shouldn't happen; if a user were to run a profile only running the vertical beam, Dolfyn would have errored out.

[jmcvey3]

Reading the stored "amp_bt" and "corr_bt" variables will massively increase the file size, as they are stored with a bin size of around 0.1 m. If someone requests this information, I can add it later.