Avoid infinite log resending if exceptions happen during the log-sending process

[amidofu]

Issue #, if available:

Description of changes:

1. We drop the logs in the log batch if the current log batch causes exceptions in the log-sending process
2. we set the auth region by the config region if the service url is used

For 1., we did this because we observed that our log batch causes the over 24 hours spanning problem
and since the monitor is not stopped, it just resends the log batch forever.

For 2., we did this because we got a problem with the auth region setup

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.

[GarrettBeatty]

do we know if this issue is still happening in the latest version @amidofu

[afroz429]

@amidofu are you able to confirm if this issue is still happening?

[amidofu]

@afroz429
sorry, I can't confirm it now, I have some trouble making your latest version work in my env, I'm trying to figure out why.

I think you can test it by adding an event like

InputLogEvent tmp = new InputLogEvent
{
    Message = "test",
    Timestamp = DateTime.UtcNow.AddDays(-2);
}
_repo.AddMessage(tmp);

to here

aws-logging-dotnet/src/AWS.Logger.Core/Core/AWSLoggerCore.cs

Line 404 in cf75428

and see if Monitor keeps sending the same log events in _repo

thank you