pd-xsample 0.3.2+git20170905.1.4441ae5-6 fails to build on ppc64el

pd-xsample 0.3.2+git20170905.1.4441ae5-6 fails to build on ppc64el:

g++ -DPD -I "/usr/include/pd" -DPD_FLOATSIZE=64 -DUNIX -Wdate-time -D_FORTIFY_SOURCE=2 -fPIC -fcheck-new -DNDEBUG -I/usr/include/flext -DFLEXT_SYS=2 -DFLEXT_SHARED -DFLEXT_USE_CMEM -pthread -I/usr/include/pd -DPD -Wdate-time -D_FORTIFY_SOURCE=2 -g -O3 -ffile-prefix-map=/<<PKGBUILDDIR>>=. -flto=auto -ffat-lto-objects -fstack-protector-strong -Wformat -Werror=format-security -fdebug-prefix-map=/<<PKGBUILDDIR>>=/usr/src/pd-xsample-0.3.2+git20170905.1.4441ae5-6 -o source/record.linux-ppc64l-64.so.o -c source/record.cpp
[...]
source/groove.cpp:358:16: error: invalid parameter combination for AltiVec intrinsic ‘__builtin_vec_dst’
  358 | vec_dst(speed,GetPrefetchConstant(1,n>>2,0),0);
      | ^

Full log: https://launchpadlibrarian.net/683921963/buildlog_ubuntu-mantic-ppc64el.pd-xsample_0.3.2+git20170905.1.4441ae5-6_BUILDING.txt.gz

Setting `PD_FLOATSIZE=64` will lead to the `speed` type `t_sample` being defined as `double`. According to https://gcc.gnu.org/onlinedocs/gcc/PowerPC-AltiVec-Built-in-Functions-on-ISA-2_002e05.html vec_dst is defined for `float` but not for `double`.

"Build Pd64 externals" was the only change of pd-xsample 0.3.2+git20170905.1.4441ae5-6. So let's skip this change for ppc64el.