AP_HAL: Use atomic instead volatile on RingBuffer head/tail

Volatile will provide protection to sequence re-ordering and guarantee
the variable is fetched from memory, but it won't provide the memory
barrier needed to ensure that no re-ordering (by either the compiler or
the CPU) will happen among other threads of execution
accessing the same variables.

For more info about this effect can be found on articles about
std::memory_order.

libraries/AP_HAL/utility/RingBuffer.cpp

@@ -10,7 +10,6 @@ ByteBuffer::ByteBuffer(uint32_t _size)
 {
     size = _size;
     buf = new uint8_t[size];
-    head = tail = 0;
 }
 
 ByteBuffer::~ByteBuffer(void)

libraries/AP_HAL/utility/RingBuffer.h

@@ -1,5 +1,6 @@
 #pragma once
 
+#include <atomic>
 #include <stdint.h>
 #include <stdbool.h>
 
@@ -103,8 +104,8 @@ private:
 
     // head is where the next available data is. tail is where new
     // data is written
-    volatile uint32_t head = 0;
-    volatile uint32_t tail = 0;
+    std::atomic<uint32_t> head{0};
+    std::atomic<uint32_t> tail{0};
 };
 
 /*