Add mult_frac() to handle fractional multiplication safely

Introduce a new inline function mult_frac() to compute "x * n / d" with
enhanced precision and safety. This function addresses the common
issues of overflow and loss of precision that can occur with
straightforward approaches to this calculation.

Directly computing "x * n / d" can lead to problems:
- Performing the division first can result in loss of precision due to
  integer truncation.
- Performing multiplication before division can risk overflow.

The mult_frac() function mitigates these issues by:
1. Calculating the quotient and remainder of 'x' divided by 'd'.
2. Using these intermediate results to perform the final calculation,
   thus avoiding intermediate overflow and preserving precision.

This approach is based on the Linux kernel's mult_frac() macro [1],
which follows a similar method to handle these calculations robustly.

Link: https://elixir.bootlin.com/linux/v6.10.7/source/include/linux/math.h#L121 [1]

Author: visitorckw

src/utils.c

@@ -26,6 +26,19 @@
 
 #define MAX_PATH_LEN 1024
 
+/* Calculate "x * n / d" without unnecessary overflow or loss of precision.
+ *
+ * Reference:
+ * https://elixir.bootlin.com/linux/v6.10.7/source/include/linux/math.h#L121
+ */
+static inline uint64_t mult_frac(uint64_t x, uint64_t n, uint64_t d)
+{
+    const uint64_t q = x / d;
+    const uint64_t r = x % d;
+
+    return q * n + r * n / d;
+}
+
 static void get_time_info(int32_t *tv_sec, int32_t *tv_nsec)
 {
 #if defined(HAVE_POSIX_TIMER)