Fix for Broken Arduino libraries due to implementaion of commonly used GPIO constants, issue #98

[frinkahedron]

For your review, a potential fix for issue #98.
The changes add a new pinMode function that maps commonly used Arduino pin mode constants to internal methods using a pincfg structure and makes changes to existing defines to reflect this update.

Please Note:

1. In the implementation of the pinMode function I used a variable to assign the desired pin mode mapping and just called the internal function once. I did this instead of calling the function at each case (without the variable) because it seemed like it looked cleaner and may optimize better. However this assumes the am_hal_gpio_pincfg_t structure is safe to assign to a new variable (does not use pointers, etc). In this case the implementation is safe but future modifications could potentially break this if the = operator is not overloaded to account for it.
   I could easily change it back to an alternate implementation that would be safer long term if you all are ok with the extra instances of functions calls in the generated code. I saw other instances in the code where assignment was being made between am_hal_gpio_pincfg_t variables so I assumed this was already factored into consideration.

2. In the implementation of the pinMode function when there is an invalid value passed in for the pin mode I just return and do nothing. There did not seem to be an error checking semantics in place for the original pinMode functions so it seemed safest just to do nothing.

3. I was trying to hold back my OCD and was pretty sparse in my comments using the existing level of detail as a guide. If a different level of detail is desired just let me know.

[jgromes]

Hi, any chance this could be added? The pin mode issue really does break a lot of Arduino libraries (including mine :P).

It's a shame Arduino core has no specific pin state type, but I'm afraid that design decision has been made way too long ago to do anything about it now...

[oclyke]

I apologize - I had some WIP response that must have gotten lost.

1. I agree with you judgement since it is already done that way in other locations in the core. We will just have to cross the bridge whenever (if ever) the implementation of the HAL changes.

2. Agreed

3. Thanks - with this project we've generally wanted to try a concept of only the comments that are necessary to explain a general concept with details from commits serving as additional resources. So what you've done is perfect!

Thanks for helping make the core better!

[jgromes]

@oclyke Perfect! Any chance we could get a release to update the boards manager? Thanks, and have a great Christmas!

[oclyke]

@jgromes Yes you may! I'm also including some addtl. improvements to the SVL bootloader. Speed and reliability are both increased thanks to Nate Seidle.