fix: low entropy calc when using Mnemonic XOR

[qlrd]

What is this PR for?

This commit add a fix for shannon entropy calculation of a choosen entroypy bytes before operate the XOR between input and current mnemonic.

Changes made to:

• Code
• Tests
• Docs
• CHANGELOG

Did you build the code and tested on device?

• Yes, build and tested on

What is the purpose of this pull request?

• Bug fix
• New feature
• Docs update
• Other

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 97.40%. Comparing base (573de43) to head (eb52af1).
⚠️ Report is 1 commits behind head on develop.

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #745      +/-   ##
===========================================
+ Coverage    97.38%   97.40%   +0.01%     
===========================================
  Files           83       83              
  Lines        10548    10586      +38     
===========================================
+ Hits         10272    10311      +39     
+ Misses         276      275       -1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[odudex]

Shannon of derivatives (variation), instead of raw data, was used in dice rolls to detect arithmetic progression patterns, but to have Shannon measurement you shouldn't derive the data.
If you're going to measure it byte-wise, the "histogram" or bucket will range from 0 to 255. But in the context of mnemonics, to have a entropy measurement maybe a 11bit words should be used, with 2048 possible values for the bucket.

[qlrd]

Tested with device and the mnemonics shared by @jdlcdl on #727.

[qlrd]

There's another mnemonic that have low entropy (2.29 IDK which unit) on tests and not documented AFAIK.

[odudex]

I liked the mnemonic entropy measurement.
But IMO, if we're going to create a setting with a threshold, we should use the measurement when loading any type of mnemonic and show a warning in case the threshold is not met.

[tadeubas]

IMHO we should not offer a settings for this, especially a default that "value was a random choice". Instead we need to ensure a "secure" value, as we already do with dice rolls and camera.

[odudex]

Yes

IMHO we should not offer a settings for this, especially a default that "value was a random choice". Instead we need to ensure a "secure" value, as we already do with dice rolls and camera.

Yes, this needs to be well discussed. The safest approach is to set hard-coded value that cover obvious problematic entropy, and don't overwhelm user with settings in this case.

[tadeubas]

It would be nice to somehow show the distribution, if possible - similar to how we do with dice - or at least display the entropy in bits instead of an error with 0.413817, since it’s not clear what that number represents.

[qlrd]

I liked the mnemonic entropy measurement. But IMO, if we're going to create a setting with a threshold, we should use the measurement when loading any type of mnemonic and show a warning in case the threshold is not met.

I thought in a bits/word unit. It's not the mnemonic word, but instead the 11bit word. Maybe this could lead to a misconsception, but write it correctly on firmware could be too big.

[qlrd]

Didn't tested on simulator and in device with new changes about warning entropy measure in both simulator and device.

In Simulator, IDK, this is happening:

❯ poetry run poe simulator
Poe => python simulator/simulator.py --device maixpy_amigo
pygame 2.6.1 (SDL 2.32.10, Python 3.14.0)
Hello from the pygame community. https://www.pygame.org/contribute.html
Traceback (most recent call last):
  File "/Users/qlrd/github/krux/simulator/simulator.py", line 49, in <module>
    parser.add_argument(
    ~~~~~~~~~~~~~~~~~~~^
        "--printer",
        ^^^^^^^^^^^^
    ...<3 lines>...
        action=argparse.BooleanOptionalAction,
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    )
    ^
  File "/opt/homebrew/Cellar/python@3.14/3.14.0_1/Frameworks/Python.framework/Versions/3.14/lib/python3.14/argparse.py", line 1538, in add_argument
    action = action_class(**kwargs)
TypeError: BooleanOptionalAction.__init__() got an unexpected keyword argument 'type'

Didn't re-build in device because wasn't able to simulate.

[tadeubas]

Did you merge with develop?

[qlrd]

Maybe use py3.14 could be the issue?

[qlrd]

Didn't tested on simulator and in device with new changes about warning entropy measure in both simulator and device.

In Simulator, IDK, this is happening:

❯ poetry run poe simulator
Poe => python simulator/simulator.py --device maixpy_amigo
pygame 2.6.1 (SDL 2.32.10, Python 3.14.0)
Hello from the pygame community. https://www.pygame.org/contribute.html
Traceback (most recent call last):
  File "/Users/qlrd/github/krux/simulator/simulator.py", line 49, in <module>
    parser.add_argument(
    ~~~~~~~~~~~~~~~~~~~^
        "--printer",
        ^^^^^^^^^^^^
    ...<3 lines>...
        action=argparse.BooleanOptionalAction,
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    )
    ^
  File "/opt/homebrew/Cellar/python@3.14/3.14.0_1/Frameworks/Python.framework/Versions/3.14/lib/python3.14/argparse.py", line 1538, in add_argument
    action = action_class(**kwargs)
TypeError: BooleanOptionalAction.__init__() got an unexpected keyword argument 'type'

Didn't re-build in device because wasn't able to simulate.

Fixed and rebased.