Remove Info from plaintext Hybrid Reports

[tyurek]

• Remove HybridInfo from being included in plaintext HybridReports by default.
• HybridInfo is now a required argument for encryption
• Because we want the info in some contexts, adds a function to optionally return both the report and the info when decrypting
• Adds a create_hybrid_info() function to TestHybridRecord so that we can do the initial encryption
• Adjusts encryption/decryption scripts to handle refactor. Output format is unchanged.

[codecov]

Codecov Report

Attention: Patch coverage is 95.69892% with 8 lines in your changes missing coverage. Please review.

Project coverage is 93.31%. Comparing base (bb543f9) to head (2df795f).
Report is 7 commits behind head on main.

Files with missing lines	Patch %	Lines
ipa-core/src/report/hybrid.rs	91.01%	8 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1487      +/-   ##
==========================================
+ Coverage   93.28%   93.31%   +0.03%     
==========================================
  Files         239      237       -2     
  Lines       43532    43599      +67     
==========================================
+ Hits        40608    40686      +78     
+ Misses       2924     2913      -11     

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

[tyurek]

probably some room for improvement here. What happens here is that we have three different reports (generated by share()), which all are encrypted with the same HybridInfo. I changed the encrypt_share function to take an info argument, but it really only needs a reference (I tried having it take a reference but ran into issues with lifetimes). Is it worth trying to use a smart pointer here? Or maybe rewrite encrypt_share to take three at a time but only one Info?

[akoshelev]

Yea it feels like using references here is the right thing to do. But lifetimes can be tricky to handle. I can take a look at it later today

[akoshelev]

One benefit of using usize here is that the caller cannot ignore the fact that report can be more than 65k. I thought we only need to do this cast once, when we encode the report length, so this module may not need to concern itself with such limitations

[tyurek]

Looking to wrap this one up. I put out an example commit here: 13fafea which modifies the encryptor to encrypt three shares per thread and use the same aad info for each. I could do either this or use a smart pointer, but doing it with lifetimes seems complicated.