CA1805: Fix FP with parameterless ctor on value types

[Evangelink]

Fixes #5750 and fixes #5887

[mavasani]

@stephentoub for review

[stephentoub]

Thanks.

[sharwell]

Waiting for question to be answered

[codecov]

Codecov Report

Merging #5813 (7a4878f) into main (58db3ef) will increase coverage by 0.01%.
The diff coverage is 98.90%.

❗ Current head 7a4878f differs from pull request most recent head 5605d64. Consider uploading reports for the commit 5605d64 to get more accurate results

@@            Coverage Diff             @@
##             main    #5813      +/-   ##
==========================================
+ Coverage   96.03%   96.04%   +0.01%     
==========================================
  Files        1338     1325      -13     
  Lines      307627   306254    -1373     
  Branches     9803     9693     -110     
==========================================
- Hits       295427   294142    -1285     
+ Misses       9817     9775      -42     
+ Partials     2383     2337      -46     

[Youssef1313]

Can you add few tests for #5887?

public record struct MyRecord()
{
    public bool SomeBool { get; set; } = false;
}

---

public record struct MyRecord
{
    public MyRecord() { }
    public bool SomeBool { get; set; } = false;
}

[Youssef1313]

There is no test for a struct without explicit constructor. This now produces CS8983 error anyway, so we may not care, but it didn't use to produce an error before dotnet/roslyn@3e4cc4c (even before that commit I'm not sure why issue states it being a false positive)

[Evangelink]

@stephentoub @Youssef1313 ready for review. Thanks!

[stephentoub]

Thanks.

[Evangelink]

@stephentoub @Youssef1313 PR should be finally ready.

[ManickaP]

I'm hitting this problem in my PR, so it would be great if we could merge this. @Evangelink is the PR ready to be merged? Has it become stale?

[Evangelink]

@ManickaP I will merge main to have the CI issues fixed but as far as I remember it was ready to be merged.