enable misra-c2012-2.3 & misra-c2012-2.4

[bongbui321]

For #1794, I included both the 2.3 and 2.4 since they are the same fix.

1. Inline suppress the CANPacket_t type because it is used in other files and it is a false positives of cppcheck to check only if it is used in that specific file - board/can_definitions.h.
2. Commented out USB_OTG_HostPortTypeDef type because they are commented out in board/drivers/usb.h and not used everywhere else

[adeebshihadeh]

hmm, why do we have to do this here but not for health.h?

[bongbui321]

yeah it was a bit tricky to understand for me in the beginning too.

if you try to run my code, cppcheck is flagging misra-c2012 2.3 at the board/pedal/main.c file only not the other two test of board/main.c. I'm not entirely sure of the implementation of cppcheck, but I read through the code base and see that they would read through all of the include header files. For instance to check for misra_2.3 it would look through all of the typedef struct definition of the main file as well as the included files. In the board/pedal/main.c, if you follow the trail then it would include some header files that include the can_definitions.h file, while the pedal/main.c file doesn't use any header file that has health.h file in it.

I also did one experiment and put #include health.h in the board/pedal/main.c file and get this.

/home/openpilot/panda/board/health.h:35:1: style: misra violation (use --rule-texts=<file> to get proper output) [misra-c2012-2.3]
typedef struct __attribute__((packed)) {
^
/home/openpilot/panda/board/health.h:4:41: style: misra violation (use --rule-texts=<file> to get proper output) [misra-c2012-2.4]
struct __attribute__((packed)) health_t {

So I think it is a false positive of cppcheck.

[adeebshihadeh]

it's not necessary, I just reverted it. please be more careful with changes like this in the future