Report faulty share not just general error

[jaromil]

On failure to combine shares, there is no indication on which is the broken share (if only one).

An additional procedure may be triggered to try indicate which one is the faulty one, giving back more information to the user. If one share is faulty, the error can be isolated by changing the system to hold an extra share server-side and use it through a cycle of retries. If more than one share is faulty then the specific share cannot be isolated. Or there may be more strategies for this?

[jaromil]

Great response from secretshare's author here timtiemens/secretshare#5

Plan is to implement support for the -paranoid:

• checksum integer sequences to detect errors
• experiment and add option for storage of the "extra share" to cycle through
• eventually include checksum for sequences of shares