feat: cache attribute symbols

[TimothyMakkison]

Description
See #475

Add a Dictionary<Type, INamedTypeSymbol?> cache to WellKnownTypes to optimize AttributeDataAccessor.

Benchmarks
Benefits mappers with many mapping methods, saves 8ms for LargeCompile and saved 300 KB of memory. Doesn't look like Compile is effected.

Old:

Method	Mean	Error	StdDev	Gen0	Gen1	Allocated
Compile	1.563 ms	0.0261 ms	0.0244 ms	82.0313	19.5313	169.32 KB
LargeCompile	59.450 ms	1.1351 ms	1.1148 ms	1428.5714	285.7143	9280.05 KB

Method	Mean	Error	StdDev	Gen0	Gen1	Allocated
Compile	1.536 ms	0.0238 ms	0.0211 ms	82.0313	19.5313	170.23 KB
LargeCompile	57.525 ms	0.5249 ms	0.5155 ms	1428.5714	285.7143	9276.68 KB

Method	Mean	Error	StdDev	Gen0	Gen1	Allocated
Compile	1.571 ms	0.0240 ms	0.0225 ms	78.1250	15.6250	169.85 KB
LargeCompile	58.310 ms	0.9858 ms	1.2819 ms	1428.5714	285.7143	9272.75 KB

New:

Method	Mean	Error	StdDev	Gen0	Gen1	Allocated
Compile	1.594 ms	0.0303 ms	0.0384 ms	78.1250	15.6250	168.44 KB
LargeCompile	51.705 ms	0.8099 ms	1.1616 ms	1375.0000	375.0000	8934.55 KB

Method	Mean	Error	StdDev	Gen0	Gen1	Allocated
Compile	1.575 ms	0.0284 ms	0.0304 ms	89.8438	23.4375	168.49 KB
LargeCompile	51.513 ms	0.7022 ms	1.1924 ms	1333.3333	333.3333	8965.35 KB

Method	Mean	Error	StdDev	Gen0	Gen1	Allocated
Compile	1.559 ms	0.0296 ms	0.0277 ms	78.1250	19.5313	169.52 KB
LargeCompile	51.809 ms	0.9939 ms	1.0207 ms	1375.0000	375.0000	8925.2 KB

[codecov]

Codecov Report

Merging #478 (5ab47d0) into main (b17f666) will increase coverage by 0.39%.
The diff coverage is 95.41%.

❗ Current head 5ab47d0 differs from pull request most recent head beb7a53. Consider uploading reports for the commit beb7a53 to get more accurate results

@@            Coverage Diff             @@
##             main     #478      +/-   ##
==========================================
+ Coverage   92.08%   92.48%   +0.39%     
==========================================
  Files         135      139       +4     
  Lines        4282     4455     +173     
  Branches      591      569      -22     
==========================================
+ Hits         3943     4120     +177     
+ Misses        225      221       -4     
  Partials      114      114              

Impacted Files	Coverage Δ
...scriptors/InlineExpressionMappingBuilderContext.cs	95.12% <ø> (ø)
...appingBodyBuilders/UserMethodMappingBodyBuilder.cs	90.47% <ø> (ø)
...Riok.Mapperly/Descriptors/MappingBuilderContext.cs	98.14% <ø> (ø)
...erly/Descriptors/MappingBuilders/MappingBuilder.cs	100.00% <ø> (ø)
src/Riok.Mapperly/Descriptors/MappingCollection.cs	100.00% <ø> (ø)
...iok.Mapperly/Descriptors/Mappings/MethodMapping.cs	95.00% <ø> (ø)
...Mappings/UserDefinedExistingTargetMethodMapping.cs	87.17% <ø> (ø)
...serMappings/UserDefinedNewInstanceMethodMapping.cs	100.00% <ø> (ø)
...pings/UserMappings/UserImplementedMethodMapping.cs	90.90% <ø> (ø)
...riptors/MappingBuilders/QueryableMappingBuilder.cs	68.42% <50.00%> (ø)
... and 29 more

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[latonz]

Thank you for this improvement 😊 The performance improvement is dependent on how many times attributes where used in the mapper declaration, therefore mappers with more attributes will benefit a lot more.

[TimothyMakkison]

Removed all fields from WellKnownTypes, removed nullables and added TryGetTypeSymbol for DateOnly and TimeOnly. WellKnownTypes is now instantiated in MapperGenerator and shared between instances saving another 400KB for LargeCompile.

The DI passes although I'm a little concerned I've broken .NET Framework by removing some of the null checks.

Ran the Benchmarks a couple of times, both appear to be slower than the previous benchmarks. I wonder if the one of the more recent versions of mapperly have made it slower.

Old:

Method	Mean	Error	StdDev	Gen0	Gen1	Allocated
Compile	1.544 ms	0.0306 ms	0.0301 ms	78.1250	15.6250	170.25 KB
LargeCompile	59.269 ms	1.0096 ms	1.7144 ms	1500.0000	500.0000	9276.7 KB

Method	Mean	Error	StdDev	Median	Gen0	Gen1	Allocated
Compile	1.564 ms	0.0301 ms	0.0358 ms	1.554 ms	82.0313	15.6250	168.29 KB
LargeCompile	60.973 ms	1.2988 ms	3.5774 ms	59.640 ms	1428.5714	428.5714	9274.68 KB

Method	Mean	Error	StdDev	Gen0	Gen1	Allocated
Compile	1.553 ms	0.0285 ms	0.0267 ms	89.8438	19.5313	172.33 KB
LargeCompile	58.498 ms	0.9764 ms	1.1991 ms	1428.5714	285.7143	9266.14 KB

New:

Method	Mean	Error	StdDev	Median	Gen0	Gen1	Allocated
Compile	1.852 ms	0.0990 ms	0.2856 ms	1.721 ms	78.1250	19.5313	171.06 KB
LargeCompile	48.584 ms	0.9688 ms	0.8589 ms	48.693 ms	1333.3333	333.3333	8531.57 KB

Method	Mean	Error	StdDev	Median	Gen0	Gen1	Allocated
Compile	2.059 ms	0.1961 ms	0.5783 ms	1.740 ms	70.3125	15.6250	168.78 KB
LargeCompile	63.516 ms	6.1086 ms	18.0113 ms	57.057 ms	1285.7143	285.7143	8532.13 KB

Method	Mean	Error	StdDev	Median	Gen0	Gen1	Allocated
Compile	2.168 ms	0.1911 ms	0.5634 ms	1.989 ms	85.9375	15.6250	170.77 KB
LargeCompile	49.377 ms	0.9812 ms	2.0482 ms	49.311 ms	1333.3333	500.0000	8534.1 KB

[TimothyMakkison]

Thanks for the review! I've moved the properties, resued the logic between Get and TryGet and used the generic overload where possible.

[github-actions]

🎉 This PR is included in version 2.9.0-next.2 🎉

The release is available on:

• GitHub release
• v2.9.0-next.2

Your semantic-release bot 📦🚀

[github-actions]

🎉 This PR is included in version 3.0.0 🎉

The release is available on:

• GitHub release
• v3.0.0

Your semantic-release bot 📦🚀