arm32: loopfilter: NEON implementation of loopfilter for 16 bpc

This operates on 4 pixels as a time, while the arm64 version
operated on 8 pixels at a time.

As the registers only fit one single 4 pixel wide slice (with one
single set of input parameters and mask bits), the high level
logic for calculating those input parameters is done with GPRs
and scalar instructions instead of SIMD as in the other implementations.
34 jobs for arm32-loopfilter16 in 4 minutes and 47 seconds (queued for 3 seconds)
Status Job ID Name Coverage
  Style
passed #503091
amd64 docker
style-check

00:00:16

passed #503092
amd64 docker
x86inc-check

00:00:16

 
  Build
passed #503103
amd64 docker
build-android-aarch64

00:00:13

passed #503102
amd64 docker
build-android-armv7

00:00:13

passed #503093
amd64 avx2 docker
build-debian

00:00:22

passed #503104
aarch64 docker
build-debian-aarch64

00:00:26

passed #503105
aarch64 docker
build-debian-aarch64-clang-5

00:00:26

passed #503108
armv7 docker
build-debian-armv7

00:00:44

passed #503109
armv7 docker
build-debian-armv7-clang-5

00:00:39

passed #503096
amd64 docker
build-debian-examples

00:00:25

passed #503111
ppc64le docker
build-debian-ppc64le

00:01:53

passed #503094
amd64 docker
build-debian-static

00:00:21

passed #503107
aarch64 docker
build-debian-werror

00:00:21

passed #503095
amd64 docker
build-debian32

00:00:22

passed #503106
macos
build-macos

00:00:51

passed #503112
amd64 docker
build-pages

00:00:14

passed #503110
amd64 docker
build-ubuntu-snap

00:00:20

passed #503100
amd64 docker
build-win-arm32

00:00:22

passed #503101
amd64 docker
build-win-arm64

00:00:15

passed #503097
amd64 docker
build-win32

00:00:40

passed #503098
amd64 docker
build-win32-unaligned-stack

00:00:36

passed #503099
amd64 docker
build-win64

00:00:42

 
  Test
passed #503113
amd64 docker
test-debian

00:00:46

97.8919%
passed #503122
aarch64 docker
test-debian-aarch64

00:00:38

passed #503124
armv7 docker
test-debian-armv7-clang-5

00:00:48

passed #503118
amd64 docker
test-debian-asan

00:01:48

passed #503114
avx2 amd64 docker
test-debian-asm

00:01:02

passed #503119
amd64 docker
test-debian-msan

00:00:41

passed #503116
amd64 docker
test-debian-mt

00:00:34

passed #503123
ppc64le docker
test-debian-ppc64le

00:02:31

passed #503120
amd64 docker
test-debian-ubsan

00:00:53

passed #503117
amd64 avx2 docker
test-debian-unaligned-stack

00:00:44

passed #503115
avx2 amd64 docker
test-debian32-asm

00:00:52

passed #503121
amd64 avx2 docker
test-win64

00:00:54