High performance logging (NET6+)

[zarusz]

• Added source generated logging via the [LoggerMessage] attribute to the NET6, NET8 targets.
• This PR covers layers:
  • SlimMessageBus.Host
  • SlimMessageBus.Host.AspNetCore
  • SlimMessageBus.Host.Memory

Related to #335

Some stats are captured using the SlimMessageBus.Host.Memory.Benchmark on my machine.

BenchmarkDotNet v0.14.0, Windows 11 (10.0.26100.2605)
12th Gen Intel Core i7-1260P, 1 CPU, 16 logical and 12 physical cores
.NET SDK 9.0.100
  [Host]     : .NET 8.0.11 (8.0.1124.51707), X64 RyuJIT AVX2
  Job-SXUBYX : .NET 8.0.11 (8.0.1124.51707), X64 RyuJIT AVX2

MaxIterationCount=30  MaxWarmupIterationCount=10

Before the changes:

Type	Method	messageCount	createMessageScope	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
PubSubBenchmark	PubSub	1000000	False	717.6 ms	12.65 ms	15.53 ms	121000.0000	3000.0000	3000.0000	1.06 GB
PubSubWithConsumerInterceptorBenchmark	PubSubWithConsumerInterceptor	1000000	False	818.6 ms	15.65 ms	20.89 ms	148000.0000	3000.0000	3000.0000	1.3 GB
PubSubWithProducerInterceptorBenchmark	PubSubWithProducerInterceptor	1000000	False	833.4 ms	13.25 ms	12.39 ms	158000.0000	3000.0000	3000.0000	1.39 GB
PubSubWithPublishInterceptorBenchmark	PubSubWithPublishInterceptor	1000000	False	840.1 ms	15.11 ms	13.39 ms	158000.0000	3000.0000	3000.0000	1.39 GB
PubSubBenchmark	PubSub	1000000	True	771.5 ms	8.56 ms	7.59 ms	138000.0000	3000.0000	3000.0000	1.22 GB
PubSubWithConsumerInterceptorBenchmark	PubSubWithConsumerInterceptor	1000000	True	939.7 ms	18.38 ms	23.90 ms	166000.0000	3000.0000	3000.0000	1.45 GB
PubSubWithProducerInterceptorBenchmark	PubSubWithProducerInterceptor	1000000	True	918.3 ms	8.86 ms	6.92 ms	176000.0000	3000.0000	3000.0000	1.54 GB
PubSubWithPublishInterceptorBenchmark	PubSubWithPublishInterceptor	1000000	True	935.9 ms	7.61 ms	6.75 ms	176000.0000	3000.0000	3000.0000	1.54 GB

After the changes:

Type	Method	messageCount	createMessageScope	Mean	Error	StdDev	Gen0	Gen1	Gen2	Allocated
PubSubBenchmark	PubSub	1000000	False	651.3 ms	11.20 ms	11.51 ms	116000.0000	3000.0000	3000.0000	1.02 GB
PubSubWithConsumerInterceptorBenchmark	PubSubWithConsumerInterceptor	1000000	False	729.8 ms	14.52 ms	12.12 ms	144000.0000	3000.0000	3000.0000	1.26 GB
PubSubWithProducerInterceptorBenchmark	PubSubWithProducerInterceptor	1000000	False	759.4 ms	12.06 ms	11.28 ms	154000.0000	3000.0000	3000.0000	1.35 GB
PubSubWithPublishInterceptorBenchmark	PubSubWithPublishInterceptor	1000000	False	752.2 ms	10.63 ms	9.94 ms	154000.0000	3000.0000	3000.0000	1.35 GB
PubSubBenchmark	PubSub	1000000	True	673.1 ms	6.32 ms	5.91 ms	130000.0000	3000.0000	3000.0000	1.14 GB
PubSubWithConsumerInterceptorBenchmark	PubSubWithConsumerInterceptor	1000000	True	769.8 ms	10.16 ms	9.01 ms	157000.0000	3000.0000	3000.0000	1.38 GB
PubSubWithProducerInterceptorBenchmark	PubSubWithProducerInterceptor	1000000	True	789.1 ms	14.11 ms	12.51 ms	167000.0000	3000.0000	3000.0000	1.47 GB
PubSubWithPublishInterceptorBenchmark	PubSubWithPublishInterceptor	1000000	True	802.6 ms	9.42 ms	7.87 ms	167000.0000	3000.0000	3000.0000	1.47 GB

[zarusz]

@EtherZa when you get a moment, could you review?

[EtherZa]

Hi @zarusz. It looks good.

The only thing I can see is that you aren't specifying the access modifie for any of the delegates/net20 implementations. Not a problem in itself, but it is a deviation from the SMB coding style.

[sonarqubecloud]

Quality Gate passed

Issues
1 New issue
0 Accepted issues

Measures
0 Security Hotspots
77.5% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud