Enable to monitor batch request failure and retries

[simukappu]

Issue #, if available:
#150

Description of changes:

• Add drop_failed_records_after_batch_request_retries config param to enable you to return failed chunk to Fluentd buffer for retrying
• Increment num_errors to monitor batch request failure from "monitor_agent" or "fluent-plugin-prometheus" when drop_failed_records_after_batch_request_retries is true (default)
• Add monitor_num_of_batch_request_retries config param to enable to increment num_errors for each plugin internal retries to monitor batch request retries from "monitor_agent" or "fluent-plugin-prometheus"
• Including make sleep reliable by measuring how long we actually slept #162

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

[simukappu]

We'll close this PR once and merge included commits into master branch. We'll publish this feature as v3.4.0.rc1. Thank you!