Functional test can fail if OS_GetLocalTime and CFE_PSP_GetTime resolution is different

[skliper]

Is your feature request related to a problem? Please describe.
TestGetTime in time_current_test.c uses OS_GetLocalTime to bound the CFE_TIME_GetTime and related calls here:

cFE/modules/cfe_testcase/src/time_current_test.c

Lines 76 to 87 in 64a6a59

	OS_GetLocalTime(&start);
	Time = CFE_TIME_GetTime();
	TAI = CFE_TIME_GetTAI();
	UTC = CFE_TIME_GetUTC();
	MET = CFE_TIME_GetMET();
	METSeconds = CFE_TIME_GetMETseconds();
	METSubSeconds = CFE_TIME_GetMETsubsecs();
	STCF = CFE_TIME_GetSTCF();
	LeapSeconds = CFE_TIME_GetLeapSeconds();
	OS_GetLocalTime(&end);

OS_GetLocalTime uses clock_gettime here:
https://github.com/nasa/osal/blob/2a2defbb6b5ecc9a0f6d5af96ff73f88edc27327/src/os/portable/os-impl-posix-gettime.c#L78

but the CFE_TIME_GetTime related calls use CFE_PSP_GetTime which for VxWorks uses vxTimeBaseGet here:
https://github.com/nasa/PSP/blob/1bd7bd9508dbe7e0084fc8ad600479c5831d7660/fsw/modules/timebase_vxworks/cfe_psp_timebase_vxworks.c#L225

The issue is (at least on VxWorks) if clock_gettime resolution is less than vxTimeBaseGet the test may fail since it can show a difference of zero yet the CFE_TIME_Get* calls may show a difference so the TimeInRange test fails.

Describe the solution you'd like
Use the same utilities and stay in the CFE_TIME_SysTime_t domain by bounding with CFE_TIME_GetTime at the start and the end vs using OS_GetLocalTime. The test is really intended to show that the conversions work correctly to the resolution provided by CFE_TIME_SysTime_t, changing resolutions or references only complicates things unnecessarily leading to possible false results.

Describe alternatives you've considered
None.

Additional context
None

Requester Info
Jacob Hageman - NASA/GSFC