Prior to this commit a function compiled with the native decorator
@micropython.native would not work correctly when accessing global
variables, because the globals dict was not being set upon function entry.

This commit fixes this problem by, upon function entry, setting as the
current globals dict the globals dict context the function was defined
within, as per normal Python semantics, and as bytecode does.  Upon
function exit the original globals dict is restored.

In order to restore the globals dict when an exception is raised the native
function must guard its internals with an nlr_push/nlr_pop pair.  Because
this push/pop is relatively expensive, in both C stack usage for the
nlr_buf_t and CPU execution time, the implementation here optimises things
as much as possible.  First, the compiler keeps track of whether a function
even needs to access global variables.  Using this information the native
emitter then generates three different kinds of code:

1. no globals used, no exception handlers: no nlr handling code and no
   setting of the globals dict.

2. globals used, no exception handlers: an nlr_buf_t is allocated on the
   C stack but it is not used if the globals dict is unchanged, saving
   execution time because nlr_push/nlr_pop don't need to run.

3. function has exception handlers, may use globals: an nlr_buf_t is
   allocated and nlr_push/nlr_pop are always called.

In the end, native functions that don't access globals and don't have
exception handlers will run more efficiently than those that do.

Fixes issue #1573.

The HAL DMA functions enable SDMMC interrupts before fully resetting the
peripheral, and this can lead to a DTIMEOUT IRQ during the initialisation
of the DMA transfer, which then clears out the DMA state and leads to the
read/write not working at all.  The DTIMEOUT is there from previous SDMMC
DMA transfers, even those that succeeded, and is of duration ~180 seconds,
which is 0xffffffff / 24MHz (default DTIMER value, and clock of
peripheral).

To work around this issue, fully reset the SDMMC peripheral before calling
the HAL SD DMA functions.

Fixes issue #4110.

Since mbedtls 2.7.0 new digest functions were introduced with a "_ret"
suffix to allow the functions to return an error message (eg, if the
underlying hardware acceleration failed).  These new functions must be used
instead of the old ones to prevent deprecation warnings, or link errors for
missing functions, depending on the mbedtls configuration.

The flash-IRQ handler is used to flush the storage cache, ie write
outstanding block data from RAM to flash.  This is triggered by a timeout,
or by a direct call to flush all storage caches.

Prior to this commit, a timeout could trigger the cache flushing to occur
during the execution of a read/write to external SPI flash storage.  In
such a case the storage subsystem would break down.

SPI storage transfers are already protected against USB IRQs, so by
changing the priority of the flash IRQ to that of the USB IRQ (what is
done in this commit) the SPI transfers can be protected against any
timeouts triggering a cache flush (the cache flush would be postponed until
after the transfer finished, but note that in the case of SPI writes the
timeout is rescheduled after the transfer finishes).

The handling of internal flash sync'ing needs to be changed to directly
call flash_bdev_irq_handler() sync may be called with the IRQ priority
already raised (eg when called from a USB MSC IRQ handler).

No need to be wasteful on DMA resources.

Some DMA channels (eg for SDIO) can be used in both directions and this
patch allows such peripherals to dynamically select the DMA direction.

MCUs that have a PLLSAI can use it to generate a 48MHz clock for USB, SDIO
and RNG peripherals.  In such cases the SYSCLK is not restricted to values
that allow the system PLL to generate 48MHz, but can be any frequency.
This patch allows such configurability for F7 MCUs, allowing the SYSCLK to
be set in 2MHz increments via machine.freq().  PLLSAI will only be enabled
if needed, and consumes about 1mA extra.  This fine grained control of
frequency is useful to get accurate SPI baudrates, for example.

This makes it easy to add a custom board definition outside of the
micropython tree, keeping the micropython submodule clean and official.

If bytearray is constructed from str, a second argument of encoding is
required (in CPython), and third arg of Unicode error handling is allowed,
e.g.:

bytearray("str", "utf-8", "strict")

This is similar to bytes:

bytes("str", "utf-8", "strict")

This patch just allows to pass 2nd/3rd arguments to bytearray, but
doesn't try to validate them to not impact code size. (This is also
similar to how bytes constructor is handled, though it does a bit
more validation, e.g. check that in case of str arg, encoding argument
is passed.)

Based on tests/extmod/uhashlib_sha1.

MD5 is still widely used, and may be important in some cases for networking
interoperability, e.g. HTTP Digest authentication.

Avoids polluting the source tree, allows to build for different (sub)archs
without intermediate cleaning.

This removes the need for a separate axtls build stage, and builds all
axtls object files along with other code.  This simplifies and cleans up
the build process, automatically builds axtls when needed, and puts the
axtls object files in the correct $(BUILD) location.

The MicroPython axtls configuration file is provided in
extmod/axtls-include/config.h

To eliminate the need for symlinks which don't work on systems like
Windows.

As per ST's DfuSe specification, and following their example code.

A recent version of arm-none-eabi-gcc (8.2.0) will warn about unused packed
attributes in USB_WritePacket and USB_ReadPacket.  This patch suppresses
such warnings for this file only.

Prior to this patch tanhf(large number) would return nan due to inf/inf.

Prior to this patch tanh(large number) would return nan due to inf/inf.

This patch adds full support for unwinding jumps to the native emitter.
This means that return/break/continue can be used in try-except,
try-finally and with statements.  For code that doesn't use unwinding jumps
there is almost no overhead added to the generated code.

The native emitter keeps the current exception in a slot in its C stack
(instead of on its Python value stack), so when it catches an exception it
must explicitly clear that slot so the same exception is not reraised later
on.

Back in 8047340d basic support was added in
the VM to handle return statements within a finally block.  But it didn't
cover all cases, in particular when some finally's were active and others
inactive when the "return" was executed.

This patch adds further support for return-within-finally by correctly
managing the currently_in_except_block flag, and should fix all cases.  The
main point is that finally handlers remain on the exception stack even if
they are active (currently being executed), and the unwind return code
should only execute those finally's which are inactive.

New tests are added for the cases which now pass.