Use INLINE and include vpx_config.h instead of plain 'inline'

This fixes compilation with MSVC.

Change-Id: Iaf7d6b0a0134968a6addf315fde6d852f298db8c

vp8/common/arm/neon/loopfilter_neon.c

@@ -9,8 +9,9 @@
  */
 
 #include <arm_neon.h>
+#include "./vpx_config.h"
 
-static inline void vp8_loop_filter_neon(
+static INLINE void vp8_loop_filter_neon(
         uint8x16_t qblimit,  // flimit
         uint8x16_t qlimit,   // limit
         uint8x16_t qthresh,  // thresh

vp8/common/arm/neon/loopfiltersimplehorizontaledge_neon.c

@@ -9,8 +9,9 @@
  */
 
 #include <arm_neon.h>
+#include "./vpx_config.h"
 
-static inline void vp8_loop_filter_simple_horizontal_edge_neon(
+static INLINE void vp8_loop_filter_simple_horizontal_edge_neon(
         unsigned char *s,
         int p,
         const unsigned char *blimit) {

vp8/common/arm/neon/loopfiltersimpleverticaledge_neon.c

@@ -9,8 +9,9 @@
  */
 
 #include <arm_neon.h>
+#include "./vpx_config.h"
 
-static inline void vp8_loop_filter_simple_vertical_edge_neon(
+static INLINE void vp8_loop_filter_simple_vertical_edge_neon(
         unsigned char *s,
         int p,
         const unsigned char *blimit) {

vp8/common/arm/neon/mbloopfilter_neon.c

@@ -9,8 +9,9 @@
  */
 
 #include <arm_neon.h>
+#include "./vpx_config.h"
 
-static inline void vp8_mbloop_filter_neon(
+static INLINE void vp8_mbloop_filter_neon(
         uint8x16_t qblimit,  // mblimit
         uint8x16_t qlimit,   // limit
         uint8x16_t qthresh,  // thresh