Change the compiler's big integer library to use libTomMath

This now replaces Bill's crappy big int implementation

src/libtommath/mp_get_double.c

@@ -0,0 +1,18 @@
+#include "tommath_private.h"
+#ifdef MP_GET_DOUBLE_C
+/* LibTomMath, multiple-precision integer library -- Tom St Denis */
+/* SPDX-License-Identifier: Unlicense */
+
+double mp_get_double(const mp_int *a)
+{
+   int i;
+   double d = 0.0, fac = 1.0;
+   for (i = 0; i < MP_DIGIT_BIT; ++i) {
+      fac *= 2.0;
+   }
+   for (i = a->used; i --> 0;) {
+      d = (d * fac) + (double)a->dp[i];
+   }
+   return mp_isneg(a) ? -d : d;
+}
+#endif