Actually, there's no extra clock pulse to wait for.

The assumption of an extra clock pulse was mistakenly derived from the
fact that by the time this driver was introduced the GPIO controller of
the test machine (bcm2835) generated spurious interrupts.

Since now spurious interrupts are handled properly this can and must be
removed in order to make TX xfers work properly.

While at it, remove duplicate gpiod_direction_input(). The data gpio
must already be configured to act as input when receiving the ACK bit.

This patch is tested with the original hardware (peripherals and board)
the driver was developed on.
Signed-off-by: Danilo Krummrich <danilokrummrich@dk-develop.de>
Link: https://lore.kernel.org/r/20220215160208.34826-4-danilokrummrich@dk-develop.deSigned-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>