Fix PC test_lag_2 failure issue

[wumiaont]

Description of PR

It's found that the PC test_lag_2 tests could fail on common_setup_teardown. This routine is using parallel thread to do config_reload. The parallel thread is by default has pool of 2 threads. The config_reload() has some changes to wait 900 seconds if safe_reload flag is not set. This has 2 issues here.

1. With 2 threads pool and we have 4 cards on testing bed, will have 2 cards doing config_reload and after that's done another 2 will have change to run. This could make total wait time have to be set to > 30 min. Currently wait time is 10m.
2. Set config_reload to be true will save many times to check if all critical services are back on line than config_reload to be false which will make API wait 15m before quit

Fix is for PC common_setup_teardown to use thread pool and set number of hosts as number of threads. Thus all cards will execute config_reload at the same time.
Also when calling config_reload() set safe_reload to be true.

Summary:
Fixes # (issue)

Type of change

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

Back port request

• 202012
• 202205
• 202305
• 202311
• 202405

How did you verify/test it?

Verified on OC test bed with the fix. Test passed.

[auspham]

Hi @wumiaont, i have raised a PR here for this #15721