crypto: Log the received device keys on an encrypted olm message

[richvdh]

MSC4147 adds a device_keys property to the plaintext content of encrypted to-device events, and support for picking this out was added in #3556/#3633.

Turns out it's quite handy to log this information.

The output of this is something like:

2024-08-30T14:40:27.061681Z DEBUG receive_sync_changes:receive_to_device_event{sender="@alice:example.org" event_type="m.room.encrypted" message_id="01J6HVNZXBPF860R0TBH3H75KZ"}:handle_decrypted_to_device_event{sender_key="curve25519:CpEg1cKt2OiASokDD23L9/buqT7LAkcUEVaFXaBSCFU" event_type="m.room_key"}: matrix_sdk_crypto::machine: Received a decrypted to-device event sender_device_keys=(Some("curve25519:CpEg1cKt2OiASokDD23L9/buqT7LAkcUEVaFXaBSCFU"), Some("ed25519:aEMaB1+jf3tzIRdyNzI514m1czCra9W23JwYXho27iw"))

[codecov]

Codecov Report

Attention: Patch coverage is 87.50000% with 1 line in your changes missing coverage. Please review.

Project coverage is 84.16%. Comparing base (224292a) to head (536891a).
Report is 27 commits behind head on main.

Files with missing lines	Patch %	Lines
...rates/matrix-sdk-crypto/src/types/events/olm_v1.rs	85.71%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3919      +/-   ##
==========================================
+ Coverage   84.13%   84.16%   +0.02%     
==========================================
  Files         266      266              
  Lines       28008    28015       +7     
==========================================
+ Hits        23565    23578      +13     
+ Misses       4443     4437       -6     

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

[bnjbvr]

Approving for the Rust side, still needs some crypto review to validate it's OK to include this info in rageshakes/logs.

[andybalaam]

These keys are safe to log because they are public keys: anyone who shares a room with the user can find them proactively by making /keys/query request.

This seems potentially dangerous because they are included in the encrypted content of the event, but the MSC mentions that they are public information and could just as easily have been included in the cleartext.