DI error when deleting scheduled jobs from dashboard after upgrade

After I upgraded one of the minor versions (1.7.18 to 1.7.20), I'm getting an error when trying to delete jobs for retry. It's also happening if I try deleting the scheduled jobs. There is one caveat though - the first time during the application lifetime, the delete is successful. Every other time, it fails.

One thing to mention is that I'm not using standard DI for .NET 5 - I'm using DryIoc which is visible from the logs as well.

The error in the logs is the following:
```
at DryIoc.Throw.It(Int32 error, Object arg0, Object arg1, Object arg2, Object arg3) in /_/src/DryIoc/Container.cs:line 13568
at DryIoc.Container.ThrowIfContainerDisposed() in /_/src/DryIoc/Container.cs:line 684
at DryIoc.Container.ResolveAndCache(Int32 serviceTypeHash, Type serviceType, IfUnresolved ifUnresolved) in /_/src/DryIoc/Container.cs:line 418
at DryIoc.Container.DryIoc.IResolver.Resolve(Type serviceType, IfUnresolved ifUnresolved) in /_/src/DryIoc/Container.cs:line 356
at Microsoft.Extensions.DependencyInjection.ServiceProviderServiceExtensions.GetService[T](IServiceProvider provider)
at Hangfire.HangfireServiceCollectionExtensions.GetInternalServices(IServiceProvider provider, IBackgroundJobFactory& factory, IBackgroundJobStateChanger& stateChanger, IBackgroundJobPerformer& performer)
at Hangfire.DefaultClientManagerFactory.GetClient(JobStorage storage)
at Hangfire.Dashboard.AspNetCoreDashboardContext.GetBackgroundJobClient()
at Hangfire.Dashboard.RouteCollectionExtensions.<>c__DisplayClass3_0.b__0(DashboardContext context, String jobId)
at Hangfire.Dashboard.BatchCommandDispatcher.Dispatch(DashboardContext context)
at Hangfire.Dashboard.AspNetCoreDashboardMiddleware.Invoke(HttpContext httpContext)
at Microsoft.AspNetCore.Builder.Extensions.MapMiddleware.Invoke(HttpContext context)
at Microsoft.AspNetCore.Diagnostics.DeveloperExceptionPageMiddleware.Invoke(HttpContext context)",
"message": "code: Error.ContainerIsDisposed;
message: Container is disposed and should not be used: \"container with ambient DryIoc.AsyncExecutionFlowScopeContext with scope {Name=null}
with Rules with {TrackingDisposableTransients, ResolveIEnumerableAsLazyEnumerable, CaptureContainerDisposeStackTrace, IgnoringReuseForFuncWithArgs} and without {ThrowOnRegisteringDisposableTransient, VariantGenericTypesInResolvedCollection}
with DefaultReuse=Singleton {Lifespan=1000}
with FactorySelector=SelectLastRegisteredFactory
with Made={FactoryMethod=ConstructorWithResolvableArguments, PropertiesAndFields=, ParameterSelector=} has been DISPOSED!
Dispose stack-trace at DryIoc.Container.Dispose()
at DryIoc.Microsoft.DependencyInjection.DryIocServiceScope.Dispose()
at Microsoft.AspNetCore.Http.Features.RequestServicesFeature.DisposeAsync()
at Microsoft.AspNetCore.Http.HttpResponse.<>c.<.cctor>b__36_2(Object disposable)
at Microsoft.AspNetCore.Server.IIS.Core.IISHttpContext.FireOnCompleted()
at System.Runtime.CompilerServices.AsyncMethodBuilderCore.Start[TStateMachine](TStateMachine& stateMachine)
at Microsoft.AspNetCore.Server.IIS.Core.IISHttpContext.FireOnCompleted()
at Microsoft.AspNetCore.Server.IIS.Core.IISHttpContextOfT`1.ProcessRequestAsync()
at System.Runtime.CompilerServices.AsyncMethodBuilderCore.Start[TStateMachine](TStateMachine& stateMachine)
at Microsoft.AspNetCore.Server.IIS.Core.IISHttpContextOfT`1.ProcessRequestAsync()
at Microsoft.AspNetCore.Server.IIS.Core.IISHttpContext.HandleRequest()
at System.Runtime.CompilerServices.AsyncMethodBuilderCore.Start[TStateMachine](TStateMachine& stateMachine)
at Microsoft.AspNetCore.Server.IIS.Core.IISHttpContext.HandleRequest()
at Microsoft.AspNetCore.Server.IIS.Core.IISHttpContext.Execute()
at System.Threading.ThreadPoolWorkQueue.Dispatch()
```

I traced the error to the following commits - https://github.com/HangfireIO/Hangfire/commit/293bc41c3715fd48cd6b377f1369a79b2dcbf75f and https://github.com/HangfireIO/Hangfire/commit/07627d4261852381fc980906f48d32af2c281f1e

What I presume is happening is that `serviceProvider` that is injected in the `DefaultClientManagerFactory` will be OK the first time I try to delete one of the jobs. However, it gets disposed after the request is completed. The next time I try to delete the job, the new request is now trying to use the old, disposed `ServiceProvider` and is throwing this error.

Start by reviewing commits 293bc41c3715fd48cd6b377f1369a79b2dcbf75f and 07627d4261852381fc980906f48d32af2c281f1e, then inspect DefaultClientManagerFactory and the dashboard delete flow described in the stack trace. Reproduce repeated retry or scheduled-job deletions with DryIoc after upgrading. Done means later requests no longer attempt to resolve services from the disposed provider.

Written by the indexing model from the issue text.