Check desired Accelerometer thread configuration with Jim

[lauralindzey]

Questions:

• Currently, message queue storage is allocated 5:1:1 for waves:turbidity:light; how do we want to shift that allocation when adding the accelerometer ones? => In practice, may not matter; we currently have enough memory to store over a day of data.
• How to handle additional error bits; another field in the SBD52 message?
• There are some fields in the SBD message that I think can be ditched -- confirm this!
• Confirm reported mean/min/max can be 16 bits, not full 20 bits of accelerometer resolution. This will be simpler to implement + save bytes.
• Confirm message prioritization based on max wave height (just like SBD52 message)