Shapeless compilation for some graphs (#687)

* shapeless compilation for some graphs

* update compile benchmark

* default compile a few activations

* buffer donation

* bugfix

* shapeless fix

* update tests to work for cpu and gpu fusion

* test kwargs

* add kwargs to compile

* Recompile when python arguments change

* no compile for tanh

* some constant tests

---------

Co-authored-by: Angelos Katharopoulos <a_katharopoulos@apple.com>

mlx/backend/common/compiled.h

@@ -49,4 +49,8 @@ void print_complex_constant(std::ostream& os, const array& x) {
 
 void print_constant(std::ostream& os, const array& x);
 
+inline bool is_scalar(const array& x) {
+  return x.ndim() == 0;
+}
+
 } // namespace mlx::core