wireless/bluetooth:  Add support for IOCTL commands to perform GATT discovery and obtain the results.

include/nuttx/wireless:  Remove bogus name from structure.  This was left over from a previous change and had me confused for awhile.

wireless/bluetooth:  Remove Rx kernel thread.  The operation of dispatching messages can be performed just as well on the high and low priority work queue.  The network driver operates the low priority work queue so that also integrates better with network frame reception.

Xbee: Adds a timeout to send logic to handle case where XBee module fails to respond to a Transmit request with a Transmit Status

Approved-by: Gregory Nutt <gnutt@nuttx.org>

stm32f7 fix CONFIG_STM32F7_STM32F77X define typo

Approved-by: Gregory Nutt <gnutt@nuttx.org>

arch/arm/include/tms570, arm/src/armv7-r, and arm/src/tms570:  Adds support for the TMS570LS3137ZWT and corrects seversl ARMv7-R and TMS570 issues

arch/arm/src/stm32 and drivers/wireless/bluetooth:  Fix some buffer management errors in HCI UART upper and lower halves.

net/sixlowpan:  Fix a nasty, byte-ordering/endian-ness problem.  The root cause has been found and corrected.  And a half dozen or so bandaid fixes were reverted.

drivers/wireless/bluetooth: Default BAUD must be 115200, Fix logic in Rx callback, Rx callback must execute on work queue.  Now sends and receives one command and response.  But hangs on the next.

arch/arm/src/imxrt:  Corrects error in serial interrupts and baud setup.  Now the baseic IMXRT1050-EVK port is ready. The console is working.

arch/arm/src/stm32/:  In HCI UART, sense of RTS and CTS are inverted.  When we speak of asserting RTS we mean setting nRTS to zero.

arch/arm/src/stm32 and configs/stm32f4discovery:  Some feeble attempts to communicate with the BT860.  Not much progress yet.

net/sixlowpan: Simplify redundant check in the previous commit.  If fullmac == true, then we must also have usemac == true and so checking usemac in that context is redundant.

arch/arm/src/imxrt:  Fix problem with with CPU hanging during boot in the clockconfig function.  configs/imxrt1050-evk/scripts:  Update linker script so that the NXP boot header will be included in the build.

net/sixlowpan:  When uncompressing link local addresses, the source of the address matters.  Apparently address by byte reversed in IP addresses.  But if the source of the address is the locally assigned address, bytes are not reversed.

Revert "net/sixlowpan:  Fix a constant value used in IPv6 address decompression.  The 0x0200 value used in the address modification should be in network order."

This reverts commit e8832be8.

wireless/bluetooth:  Soften use of enter_critical_sectino() which is a performance hog in SMP configurations.  Replace with spin_lock_irqsave() in most cases.  That is the same for single core CPUs, but much more efficient in SMP configurations.  And better, repace with sched_lock() if there is really no necessity to disable interrupts too.

tools/mkromfsimg.sh:  Include an option in mkromfsimg.sh to ignore FAT/RAMDISK. This permits you to build tiny systems with ROMFS but with the rather large FAT FS support.

fs/mount:  Fix wrong size reported by df -h.  Reason is that size in bytes is calculated always on uint32_t types and cards bigger than 4GB will overflow uint32_t type. Solution is to use uint64_t where possible.

wireless/bluetooth:  Use nxsem_timedwait() vs nxsem_wait() when waiting for response from the HCI device.  Adding this timeout prevents things from hanging when there is a problem with the connect HCI UART (in the current test, when there is no HCI UART connected).

wireless/bluetooth and drivers/wireless/bluetooth:  Simply the driver registration/initialization interface.

arch/arm/src/stm32:  In HCI UART driver, need to disable Tx interrupts after frame sent.  Otherwise, we get infinite Tx interrupts.

Various fixes/changes from intial testing of the HCI UART on the STM32F4-Discovery board.  Re-organized how the Blutetooth network driver is initialized.  Added lots of new debug output.  Add a new STM32 F4 Discovery configuration for testing the HCI UART.  Various fixes to get through initialization (still does not).

fs/smartfs:  Fix a hardfault in SmartFS.  Problem occurred when opening a file with O_CREAT (only) when the file arelready exists but has a size of zero.

configs/stm32fdiscovery:  Add support in bringup logic to initialize the HCI UART and register the HCI UART network device.