[Patch, Fortran] C Binding - module+intrinsic cleanup+bug fixes

[Dominique Dhumieres]

Is it expected that the test gfortran.dg/c_loc_test_19.f90 segfaults
at runtime? The valgrind output is

==48894== Use of uninitialised value of size 8
==48894==    at 0x100000DB7: test.1891 (c_loc_test_19.f90:14)
==48894==    by 0x100000E6A: MAIN__ (c_loc_test_19.f90:9)
==48894==    by 0x100000EA2: main (c_loc_test_19.f90:9)
==48894== 
==48894== Invalid write of size 8
==48894==    at 0x100000DB7: test.1891 (c_loc_test_19.f90:14)
==48894==    by 0x100000E6A: MAIN__ (c_loc_test_19.f90:9)
==48894==    by 0x100000EA2: main (c_loc_test_19.f90:9)
==48894==  Address 0x0 is not stack'd, malloc'd or (recently) free'd
==48894== 
==48894== 
==48894== Process terminating with default action of signal 11 (SIGSEGV)
==48894==  General Protection Fault
==48894==    at 0x100293FC1: dyld_stub_binder (in /usr/lib/libSystem.B.dylib)
==48894==    by 0x10013102F: ??? (in /opt/gcc/gcc4.9w/lib/libgfortran.3.dylib)
==48894==    by 0x100000E6A: MAIN__ (c_loc_test_19.f90:9)
==48894==    by 0x100000EA2: main (c_loc_test_19.f90:9)

Dominique