Revert "prevent floating point exception signals if ECL_OPT_TRAP_SIGFPE is false"

This reverts commit 2a9084b105.

It turned out that #347 was caused by the Maxima computer algebra
system enabling floating point exceptions via si:trap-fpe (see
https://trac.sagemath.org/ticket/22191). Hence we can revert to the
less intrusive behaviour of not changing the floating point
environment if ECL_OPT_TRAP_SIGFPE is false.

CHANGELOG

@@ -114,12 +114,6 @@ in C code
   is still referenced in the memory, it's programmer duty to call wait.
 - The ECL_OPT_SIGALTSTACK_SIZE option has been removed, because it had no
   effect.
-- A false value of the ECL_OPT_TRAP_SIGFPE option now prevents floating
-  point exception signals from being generated by default. In version
-  16.1.3, ECL would not change the options controlling the generation of
-  such signals and simply not install a signal handler for floating point
-  exceptions. This could lead to such signals being generated and caught by
-  another signal handler if ECL was used as an embedded library.
 - Non-standard package nicknames (USER for COMMON-LISP-USER and LISP for
   COMMON-LISP) have been removed.
 * 16.1.3 changes since 16.1.2

src/c/unixint.d

@@ -1401,8 +1401,6 @@ install_fpe_signal_handlers()
 /*                 si_trap_fpe(@'division-by-zero', ECL_NIL); */
 /*                 si_trap_fpe(@'floating-point-overflow', ECL_NIL); */
 /* # endif */
-        } else {
-                si_trap_fpe(ECL_T, ECL_NIL);
         }
 #endif
 }