Enhance benchmarking automation

[nikhil-sk]

Is your feature request related to a problem? Please describe.

The feature requests come from comments on this PR: #1052

Describe the solution

Add the following features to automation to facilitate better user experience:
Review notes:

1. Consider using https://pypi.org/project/mdutils/ if the report structure becomes complex.
2. Consider using boto3 for aws calls?
3. Make uuid more human readable.
4. Move some config items from init.py to a config file for easy manipulation.
5. Make 'ec2 Name' tag configurable for users to easily name their instances.
6. Add useful command line arguments i.e. '--run-only '
7. call 'report.py' automatically.
8. Add appropriate logging to cloudwatch
9. Reduce duplication in test_files.

[dk19y]

Can you also log the metrics with appropriate dimensions to cloudwatch aswell.

[nikhil-sk]

Considerations addressed/not addressed/punted as of pr #1099

1. Consider using https://pypi.org/project/mdutils/ if the report structure becomes complex.
   Addressed: the 3p utility need not be used as of PR Add neuron benchmarking to automation and other enhancements #1099 as the format of the output markdown doesn't change, and is kept simple.
2. Consider using boto3 for aws calls?
   Addressed: There is no advantage to using boto3 in the current implementation. It could only lead to more verbose code. This will be dropped from list of enhancement.
3. Make uuid more human readable
   Addressed: the uuid is a unique combination of timestamp and several other factors. To make uuid more human readable, another method would need to be written. Since, the user does not directly interact with uuid, the existing implementation should serve well.
4. Move some config items from init.py to a config file for easy manipulation.: Done
5. Make 'ec2 Name' tag configurable for users to easily name their instances.: Done
6. Add useful command line arguments i.e. '--run-only ': Punted to phase 3 issue Enhance benchmarking automation phase 3 #1121
7. call 'report.py' automatically.
   Addressed: report is now generated automatically after benchmarking completes
8. Add appropriate logging to cloudwatch: Punted to phase 3 issue Enhance benchmarking automation phase 3 #1121
9. Reduce duplication in test_files.
   Addressed: Several functions were moved to a common location in PR Add neuron benchmarking to automation and other enhancements #1099 , however more can be done. and hence also included in phase 3 Enhance benchmarking automation phase 3 #1121

[chauhang]

Remaining items to be addressed in 0.5.0