BitBox02 firmware

This commit rewrites a small part of the BitBox02 U2F feature from C to Rust. When you register or authenticate with an unknown website, the device shows a short mnemonic phrase instead of a site name. The change swaps the old libwally BIP39 library for the Rust bip39 crate to generate that phrase. It is a routine refactoring/porting change; there is no direct evidence it fixes or introduces a security vulnerability, though any rewrite can carry subtle bugs.

This commit adds a new software library (a BIP-39 implementation for handling recovery seed phrases) to the BitBox02 firmware and stores a copy of it inside the project's source tree (vendoring). It does not change any device behavior by itself; it is purely a dependency addition. There is no direct evidence in the commit of a security vulnerability, but adding a new dependency always slightly increases the attack surface and supply-chain risk.

This commit is a pure test-infrastructure refactoring. It merges duplicate mock and fake hardware-implementation files that previously existed separately for unit tests and the simulator into a single shared library called 'hardware-mocks'. It also adjusts Rust build flags so RTT logging only compiles for the real firmware target (target_os='none'). There is no change to the actual device firmware or to any security-critical runtime behavior.

This commit removes an unused CMake variable and the code that built it. It is a routine cleanup with no apparent security relevance.

This commit adjusts build conditions so that a low-level hardware reset routine for the Bluetooth chip is skipped during automated unit-test builds. It is a build/test hygiene change rather than a fix for an exploitable security flaw in production firmware. There is no direct evidence that this change was made in response to a security incident or vulnerability report.

This commit is a routine code cleanup that turns the screen drawing code into a generic interface so the same code can run on both the real device and a software simulator. It does not fix or introduce any security issue.

This commit only fixes garbled comments in the source code. The actual code logic—mapping Latin-1 byte values to other byte values for display—remains completely unchanged. It is a cosmetic/documentation cleanup with no security relevance.

This commit is a routine code cleanup. It removes a thin Rust wrapper type called Bip39Wordlist and replaces it with direct use of a plain list of word indices. The behavior of the BIP39 wordlist used for entering recovery words on the device stays the same; no security vulnerability is introduced or fixed.

This commit is a routine code cleanup in the on-screen keyboard used to type BIP39 wallet recovery words. It merges two functions that both scanned the BIP39 wordlist into one loop, aiming to make typing feel faster. There is no indication in the commit or supplied references that this fixes a security bug.

This is a harmless code cleanup in the BitBox02 hardware wallet's on-screen keyboard. The developer merged two functions that did similar work into one, so the device no longer loops through the BIP39 word list twice when typing a recovery word. The visible behavior is unchanged; only performance improved.

This is a routine internal refactoring of how the on-screen keyboard looks up BIP39 seed words. It changes the code from carrying around actual word strings to carrying around word numbers and asking a helper function for the word when needed. There is no security bug being fixed here and no new attack path introduced by the change itself.

This commit changes a build-time helper tool so that, instead of crashing with a generic panic when protobuf compilation fails, it prints the actual error message and exits with a clean error code. It is a developer-experience improvement with no runtime security effect on the BitBox02 device firmware.

This commit is a performance and reliability optimization, not a typical security bug fix. It lets the BitBox02 firmware choose whether to compute an extended public key (xpub) once or twice. Computing twice helps catch rare hardware bit-flip errors when sharing xpubs with a computer. Computing once reduces workload on the secure chip during Bitcoin transaction signing, lowering the risk of hitting the chip's rate limit and causing throttling. The change is intentional and the security trade-off is documented: during signing, a bit flip would just produce an invalid signature, not leak or lose funds.

This commit is a performance optimization for the BitBox02 hardware wallet. It changes how Bitcoin transaction signing uses the secure chip, so the device only derives a private key once per transaction input instead of twice. This reduces the chance of hitting the secure chip's rate limit and avoids throttling. The change does not appear to introduce a security flaw; it is a refactoring to reuse an already-derived key for two related signing steps.