nrfs: Added explicit enum values

[Rafal-Nordic]

Added explicit enum values for better resistance to change

[karstenkoenig]

There still is a lot missing in the other ones

hal_nordic/nrfs/include/internal/services/nrfs_swext.h

Line 18 in d9f5f74

SWEXT_PD_CLAMP_DISABLED,

I think assigning fixed enums everywhere will greatly reduce the chance of accidental breakage, when enums are hardcoded people tend to make sure to not insert in the middle or reorder, when they are not it seems the order doesn't matter and one needs to be very careful during reviews to catch enums changing sequence that need to have compatibility kept in mind

[karstenkoenig]

Please double check the nrfs_temp.h change so that it didn't reorder, otherwise looks good

[karstenkoenig]

Does this need to be reordered? NRF_TEMP_EVT_REJECT assign 1 and NRFS_TEMP_EVT_CHANGE to 2 to make things not change?
I tried searching for NRFS_TEMP_SERVICE_SUBSCRIPTION_ENABLED and haven't found it, so I assume NRFS_TEMP_EVT_REJECT used to map to 1 and NRFS_TEMP_EVT_CHANGE wasn't used?

[Rafal-Nordic]

I tried searching for NRFS_TEMP_SERVICE_SUBSCRIPTION_ENABLED and haven't found it, so I assume NRFS_TEMP_EVT_REJECT used to map to 1 and NRFS_TEMP_EVT_CHANGE wasn't used?
These changes are merged in the upstream zephyr and still waiting for merging in the ncs (nrfconnect/sdk-zephyr#3050)
Current order is ok, it ensures compatibility with platforms which still have temp_subscription (nrf9230)
Explicit enum values should now fix the potential problem of using incompatible ncs/soc bundle versions. REJECT will be always interpreted as 2, regardless of whether SCFW has temp_subscription.

[karstenkoenig]

Explicit enum values should now fix the potential problem of using incompatible ncs/soc bundle versions
Nice, thanks!