crypto: Configurable sender trust checking on decrypting

[uhoreg]

Replaces #3701. Part of invisible crypto.

The room event decryption function now takes a DecryptionSetting which allows you to require that event senders have a certain trust level based on their cross-signing status. Events that do not meet the required trust level will fail to decrypt with an error.

• Public API changes documented in changelogs (optional)

Signed-off-by:

[codecov]

Codecov Report

Attention: Patch coverage is 86.95652% with 6 lines in your changes missing coverage. Please review.

Project coverage is 84.08%. Comparing base (468ee53) to head (00e079b).
Report is 81 commits behind head on main.

Files with missing lines	Patch %	Lines
crates/matrix-sdk-crypto/src/machine/mod.rs	86.66%	4 Missing ⚠️
...es/matrix-sdk-common/src/deserialized_responses.rs	71.42%	2 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #3899   +/-   ##
=======================================
  Coverage   84.07%   84.08%           
=======================================
  Files         266      266           
  Lines       27861    27894   +33     
=======================================
+ Hits        23425    23455   +30     
- Misses       4436     4439    +3     

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

[richvdh]

@uhoreg does this relate to a specific issue? Please link it if so. If not, please give more context about what we're trying to achieve in this PR.

[bnjbvr]

(Also please follow the contribution guidelines for naming PRs and commits! It helps figuring out which PR relates to what :))

[richvdh]

Could you give some more explanation about what happens to events that do not met these requirements?

[uhoreg]

I think that I've addressed all the review comments

[bnjbvr]

Can you follow the contribution guidelines for the naming of this PR and the commits, please?

[richvdh]

Looks good to me now except:

• Per #3899 (comment), #[derive(Error)] seems like an unclear way to implement Display.
• Per #3899 (comment), please update the description of the PR to describe what it is we're fixing.
• Per benji's requests, please give the PR a name that reflects the area of the code that is affected (crypto:, I guess)
• Also please rebase your commits to remove the "minor fixes" and "address review", and rename them so that they match the contributing guide. If you pick the right point on main to rebase on top of (468ee53), then github will show the force-push with no changes, which means we keep an audit trail that the code has been reviewed.

[richvdh]

🎉 🚢