Issue 81: Customize pravega payload attributes before submitting to make each payload unique

[dada-dell-emc]

Change log description
Added changes to customize pravega payload Timestamp attribute before submitting to Pravega, to make each payload unique.

Purpose of the change
Fixes #81.

What the code does
This PR does the following:
Introduced new boolean configuration attribute customPayload.
If customPayload is set to true, then customize json request payload with Timestamp dynamic value in ms.

How to verify it
Enable customPayload configuration for pravega and then run omb for pravega driver with sample payload custom-payload-idrac.data and workload custom-payload-idrac.yaml shared in this repository.
After that when you read those messages in pravega consumer check Timestamp attribute value, you will get different values almost each time in milliseconds. Or you can enble debug log in omb which will print customized payload for pravega, check Timestamp attribute value,it will print different values almost each time in milliseconds.

[ShwethaSNayak]

Did we verify these changes through any tests?

[dada-dell-emc]

Did we verify these changes through any tests?

We verified these changes from logs, there is no test cases for drivers

[sachin-j-joshi]

LGTM

[RaulGracia]

@sachin-j-joshi @dada-dell-emc do you know that the most recent version of the Pravega benchmark is in the official repository? https://github.com/openmessaging/benchmark
I say that because I don't know if the version in this repo actually has the most recent changes compared to the official repo. Besides, we contributed the Pravega driver to the official OpenMessaging repo for greater visibility and wider usage; keeping that repo updated would be great, as anybody benchmarking other system could also try Pravega with the most recent changes.

[RaulGracia]

If the intention of this PR is just to make payloads unique on a per-write basis, please consider a more efficient approach, as the current one looks quite expensive for a benchmark. Also, please note my comment about contributing in the future (maybe not this change, but future improvements such as upgrading dependencies) to the official OpenMessaging Benchmark repo, as we contributed the Pravega driver there a couple of years back: https://github.com/openmessaging/benchmark/tree/master/driver-pravega

[dada-dell-emc]

If the intention of this PR is just to make payloads unique on a per-write basis, please consider a more efficient approach, as the current one looks quite expensive for a benchmark. Also, please note my comment about contributing in the future (maybe not this change, but future improvements such as upgrading dependencies) to the official OpenMessaging Benchmark repo, as we contributed the Pravega driver there a couple of years back: https://github.com/openmessaging/benchmark/tree/master/driver-pravega

These changes are mainly just to submit the payload into pravega and not benchmarking pravega, so that QE can maintain throughput for their end to end testing.
We will improve efficiency of code for benchmark in our next PR.
New issue #82 created to address it.

[sachin-j-joshi]

@dada-dell-emc I have merged this PR based on 3 approvals. Please work towards addressing issues
in #82