Mnemonic confirmation UX

[daniel-flex]

Regarding the process for Mnemonic confirmation, after internal discussions have mostly agreed upon this from aqeel:

With sufficient context provided, forcing a check is required to provide enough friction for users to actually grasp that it's important. I would say the subset should be 6-12 leaning towards 12 and perhaps from a particular area so it is less cumbersome.

The way I had it designed originally was it picks from the 4 sub-sections so it picks two rows from 1-6, 7-12, 13-18, 19-24. For you to fill out.

And you can provide a disclaimer that we are putting the check in for demonstration purposes but still their responsbility and dont take any onus of confirming the validity for all 24 words.
putting in 12 random numbers is more difficult that it is just to enter 24 in a row. So having some sort of order helps negate that.

There's also some discussion around putting different options behind a build flag, so devs/QA can skip the process if need be.

[daniel-flex]

moving forward with this today @paulhauner. For clarification, as discussed with @qnou, will do the following:

1. user will need to confirm 1 row (6 words) from rows 1 & 2, and 1 row from rows 4 & 5.
2. confirmation screen will show all 24 slots, with some rows disabled, and some empty, ready for user input (see below)

so confirmation interface will look like:

[****][****][****][****][****][****]
[____][____][____][____][____][____]
[____][____][____][____][____][____]
[****][****][****][****][****][****]

Also, will we always force confirmation of the last word, to ensure the user has 24 words written down? (waiting on confirmation from @qnou for this)

I'll also put an evn flag in place for developers to display a 'skip' button on the confirmation page - for testing.

[daniel-flex]

pushed mnemonic confirmation type update: dde4c28
pushed skip button on 8f098db

also, new env flag for mnemonic confirmation type:
REACT_APP_MNEMONIC_CONFIRMATION_TYPE=ALL
options: ALL|TWOROWS

@qnou @paulhauner test and confirm

[paulhauner]

Tested on my end, skip button is working and so is two-row confirmation.

Something to note is that if you confirm the mnemonic, then go back you can get a different two rows to confirm. Not a deal breaker, but might be nice to avoid if the fix is simple.

[daniel-flex]

@paulhauner will set a confirmed flag and and bypass if already confirmed

[qnou]

@daniel-flex anything left remaining on this from my end?

[daniel-flex]

@paulhauner added in #2f65f9d