[Chassis][T2]: Reduce traffic-shift TSA/TSB tests runtime

[sanjair-git]

Description of PR

Summary:
Fixes # (issue)

• This PR helps to reduce the runtime of TSA/TSB tests on T2 chassis and attempts to fix issue [Bug]: New TSA/TSB tests take a long time (15+ hours) and generate a lot of logs (50+ GB) #15830

Type of change

• Bug fix
• Testbed and Framework(new/improvement)
• Test case(new/improvement)

Back port request

• 202012
• 202205
• 202305
• 202311
• 202405

Approach

What is the motivation for this PR?

• Traffic shift test cases of BGP test suite takes longer time for execution as it involves reboot of line cards or supervisor which in turn makes the tests verification mandatory for processes, neighbors, routes, tsa-tsb-service etc. before and after test case execution.
• These lead to generate lots of log for BGP traffic shift test cases, which are difficult for analysis.

How did you do it?

• Instead of executing tests on each of the line card under supervisor of different HwSKU, choose a random line card of any HwSKU for each test.
• Increase the concurrent parallel tasks from 8 to 12 for parsing the routes.

How did you verify/test it?

• From what I have observed, this reduces the test run time by half without compromising the functionality check.
• Increasing the route processing tasks from 8 to 12, reduced test case execution time by 4 minutes for each test case.

For example,

Before these changes: Startup TSA test suite - 488.72 minutes - 8.15 hrs
After these changes:  Startup TSA test suite - 234.51 minutes - 3.9 hrs

Reliable-TSA tests deal with supervisor card, hence these changes are not effective on them.

Any platform specific information?

Supported testbed topology if it's a new test case?

Documentation

[mssonicbld]

/azp run

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[Javier-Tan]

LGTM