This saves a bit in code size, and saves some precious .bss RAM:

                 .text  .bss
minimal CROSS=1: -28    -4
unix (64-bit):   -64    -8

This saves a bit in code size:

minimal CROSS=1: -44
unix:            -96

This macro is written out explicitly in the two locations that it is used
and then the code is optimised, opening possibilities for further
optimisations and reducing code size:

unix:            -48
minimal CROSS=1: -32
stm32:           -32

Based on testing, this is how the mapping should be.

Using the message "maximum recursion depth exceeded" for when the pystack
runs out of memory can be misleading because the pystack can run out for
reasons other than deep recursion (although in most cases pystack
exhaustion is probably indirectly related to deep recursion).  And it's
important to give the user more precise feedback as to the reason for the
error: if they know precisely that the pystack was exhausted then they have
a chance to increase the amount of memory available to the pystack (as
opposed to not knowing if it was the C stack or pystack that ran out).

Also, C stack exhaustion is more serious than pystack exhaustion because it
could have been that the C stack overflowed and overwrote/corrupted some
data and so the system must be restarted.  The pystack can never corrupt
data in this way so pystack exhaustion does not require a system restart.
Knowing the difference between these two cases is therefore important.

The actual exception type for pystack exhaustion remains as RuntimeError so
that programatically it behaves the same as a C stack exhaustion.

By adding __builtin_unreachable() at the end of nlr_push, we're
essentially telling the compiler that this function will never return.
When GCC LTO is in use, this means that any time nlr_push() is called
(which is often), the compiler thinks this function will never return
and thus eliminates all code following the call.

Note: I've added a 'return 0' for older GCC versions like 4.6 which
complain about not returning anything (which doesn't make sense in a
naked function). Newer GCC versions (tested 4.8, 5.4 and some others)
don't complain about this.

The machine.Pin class is also updated to support these wake-on-pin
features.

The machine.RTC class is added and the machine module is updated with the
implementation of sleep, deepsleep, reset_cause and wake_reason.

This constant exception instance was once used by m_malloc_fail() to raise
a MemoryError without allocating memory, but it was made obsolete long ago
by 3556e457.  The functionality is now
replaced by the use of mp_emergency_exception_obj which lives in the global
uPy state, and which can handle any exception type, not just MemoryError.

The formatting of exception objects is done as per CPython conventions, eg:

    :exc:`TypeError`

Prior to this patch a negative duty would lead to full PWM.

This feature is not often used so is guarded by the config option
MICROPY_PY_BUILTINS_RANGE_BINOP which is disabled by default.  With this
option disabled MicroPython will always return false when comparing two
range objects for equality (unless they are exactly the same object
instance).  This does not match CPython so if (in)equality between range
objects is needed then this option should be enabled.

Enabling this option costs between 100 and 200 bytes of code space
depending on the machine architecture.

Most embedded targets will have this bit of the code disabled, saving a
small amount of code space.

This patch provides inline versions of the utf8 helper functions for the
case when unicode is disabled (MICROPY_PY_BUILTINS_STR_UNICODE set to 0).
This saves code size.

The unichar_charlen function is also renamed to utf8_charlen to match the
other utf8 helper functions, and the signature of this function is adjusted
for consistency (const char* -> const byte*, mp_uint_t -> size_t).

Instead of putting just 'CRASH' in the .py.out file, this patch makes it so
any output from uPy that led to the crash is stored in the .py.out file, as
well as the 'CRASH' message at the end.

For builds without the compiler enabled (and hence without eval/exec) it is
useful to still be able to run as many tests as possible.

Prior to this patch, storage.c was a combination of code that handled
either internal flash or external SPI flash and exposed one of them as a
block device for the local storage.  It was also exposed to the USB MSC.

This patch splits out the flash and SPI code to separate files, which each
provide a general block-device interface (at the C level).  Then storage.c
just picks one of them to use as the local storage medium.  The aim of this
factoring is to allow to add new block devices in the future and allow for
easier configurability.

This patch allows to completely compile-out support for USB, and no-USB is
now the default.  If a board wants to enable USB it should define:

    #define MICROPY_HW_ENABLE_USB (1)

And then one or more of the following to select the USB PHY:

    #define MICROPY_HW_USB_FS (1)
    #define MICROPY_HW_USB_HS (1)
    #define MICROPY_HW_USB_HS_IN_FS (1)

Newer versions of the HAL use names which are cleaner and more
self-consistent amongst the HAL itself.  This patch switches to use those
names in most places so it is easier to update the HAL in the future.

Prior to this patch the USBD driver did not handle the recipient correctly
for setup requests.  It was not interpreting the req->wIndex field in the
right way: in some cases this field indicates the endpoint number but the
code was assuming it always indicated the interface number.

This patch fixes this.  The only noticeable change is to the MSC
interface, which should now correctly respond to the USB_REQ_CLEAR_FEATURE
request and hence unmount properly from the host when requested.

mpconfigboard_common.h now sets the defaults so there is no longer a need
to explicitly list all configuration options in a board's mpconfigboard.h
file.