This makes it a hell of a lot simpler but will lose support for BSD
make. All effected machines are running gnu-make, so the cost benefit
ratio ways heavily in favor of doing this.
It will look for all first level child `makefile`, and run assigned target. This
is equivalent to running:
```shell
for i in */makefile; do
make -C $i $target
done
```
This will case `make update` to run everytime there is a successful
merge. This will include everytime the is a non no-op merge. This *will
not run* if:
1. There is no changes.
2. There is a conflict in the merge.
Additionally the make install set will now sym link the post-merge hook.