c_src/double-conversion/utils.h: Use __arm__ instead of __ARMEL__

Big and little endian ARM both use 64bit doubles, so use the more
general __arm__ rather than __ARMEL__.

c_src/double-conversion/utils.h

@@ -53,7 +53,7 @@
 // disabled.)
 // On Linux,x86 89255e-22 != Div_double(89255.0/1e22)
 #if defined(_M_X64) || defined(__x86_64__) || \
-    defined(__ARMEL__) || defined(__avr32__) || \
+    defined(__arm__) || defined(__avr32__) || \
     defined(__hppa__) || defined(__ia64__) || \
     defined(__mips__) || defined(__powerpc__) || \
     defined(__sparc__) || defined(__sparc) || defined(__s390__) || \