Message ID support: The sending side

[poljar]

This closes: #1266

The first patch cleans some logs up, secondly we add optional support to send
message IDs for our m.room.encrypted to-device events, and finally we log the
message ID when we encrypt room keys.

• Public API changes documented in changelogs (optional)

[codecov]

Codecov Report

Patch coverage: 84.09% and project coverage change: +0.08 🎉

Comparison is base (c042e1e) 75.72% compared to head (b6bb0f3) 75.81%.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1946      +/-   ##
==========================================
+ Coverage   75.72%   75.81%   +0.08%     
==========================================
  Files         142      142              
  Lines       16271    16276       +5     
==========================================
+ Hits        12321    12339      +18     
+ Misses       3950     3937      -13     

Impacted Files	Coverage Δ
crates/matrix-sdk-crypto/src/gossiping/machine.rs	63.39% <50.00%> (+2.53%)	⬆️
...x-sdk-crypto/src/session_manager/group_sessions.rs	86.29% <79.16%> (-0.02%)	⬇️
crates/matrix-sdk-crypto/src/identities/device.rs	74.80% <100.00%> (+0.50%)	⬆️
crates/matrix-sdk-crypto/src/olm/account.rs	82.93% <100.00%> (+0.04%)	⬆️
crates/matrix-sdk-crypto/src/olm/session.rs	98.18% <100.00%> (+0.03%)	⬆️
...trix-sdk-crypto/src/types/events/room/encrypted.rs	83.05% <100.00%> (+1.56%)	⬆️

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[poljar]

And ulid doesn't work on WASM 🥲.

[bnjbvr]

Ah, makes sense; it probably requires a source of randomness or time for the wasm side that it doesn't have otherwise; it could be done via a custom import, or... compiling to the WASI target and include a WASI polyfill in the browser.