[aws-cpp-sdk-core]: increase STS reliability and retries

[grrtrr]

This fixes issues we have repeatedly experienced when using STS for authentication in a large Kubernetes cluster, with heavy load on STS:

1. The default connect timeout of 1s is too low. It happens that connections slow down. One case is very high load on kube DNS. A value of 30 seconds has proven to be robust.
2. The retry parameters are too short, authentication would frequently fail whenever STS was under higher load. The retry settings have worked in production for about 2 years.

Check all that applies:

• [X] Did a review by yourself.
• [X] Added proper tests to cover this PR. (If tests are not applicable, explain.) We have before/after experience. Without these settings, calls to STS would frequently time out whenever STS was under higher load. Within the close to 2 years after making this change, we no longer experienced authentication failures when STS was under load.
• [X] Checked if this PR is a breaking (APIs have been changed) change.
• [X] Checked if this PR will not introduce cross-platform inconsistent behavior.
• [X]] Checked if this PR would require a ReadMe/Wiki update.

Check which platforms you have built SDK on to verify the correctness of this PR.

• [X] Linux
• [ ] Windows
• [ ] Android
• [ ] MacOS
• [ ] IOS
• [ ] Other Platforms

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.