Failing UT fixes

[ghost]

Description

Two unit tests fail after applying the -fshort-enums flag.

To fix the LTC4275 update_status test, either need to change the functionality of update_status to have an else after this if or need to change expected return values that there is no else and so values should be unmodified. Went with the second option for now, so expecting values to be unchanged and compare with the values of state and pdAlert from before the update_status is called.
While I was editing this test suite, I also updated all integer expect values to be the correct enum value instead.

To fix the ocmp_ltc4275 test fail, I just increased the maximum number of GPIO pin callbacks in the GPIO stub to be 32, which is greater than the index of 27 used in the failing test. This isn't really a fix for the underlying issue, so I also added a check to the index so that future tests should fail if they index beyond FAKE_GPIO_PIN_COUNT.

Test Plan

Verify Fixes known issues
Run Unit Tests (make clean test) and verify that all tests now pass.

Verify Index Check fails correctly
Modify FAKE_GPIO_PIN_COUNT to be small (set to 1) and verify that this causes tests to fail because of returning GPIO error and not because of writing incorrect memory space. (test_ina226_alerts, test_se98a_alerts, and test_LTC4015_alerts all fail).
If running the same thing without the index check in fake_GPIO.c, the tests do not finish because of memory corruption.

Issues

Fixes #63
Fixes #113

[JoshuaJeyaraj]

To fix the ocmp_ltc4275 test fail, I just increased the maximum number of GPIO pin callbacks in the GPIO stub to be 32, which is greater than the index of 27 used in the failing test. This isn't really a fix for the underlying issue, so I also added a check to the index so that future tests should fail if they index beyond FAKE_GPIO_PIN_COUNT.

Nice catch !!
I am not sure how this issue is not showing up in the cygwin environment.
I think the max number should be increased to OC_EC_GPIOCOUNT , the rest looks good to me

[jspectacular]

coverage target does not work. e.g. make coverage fails. Not to be confused with make cov, which does work. The coverage target only works when invoked by some other rule. Otherwise, the unit test summary shows 0 total failures.