More fixes for arrays with large sizes (#1405)

* compile works for big arrays when contiguous

* style

* nits in docs

* a bunch more stuff

* update jit

* update jit

* use constant for shapes and strides and remove elem_to_loc overload

* use kernel instantiation

* docs nits

* update binary and ternary

* comments

mlx/backend/common/CMakeLists.txt

@@ -51,6 +51,7 @@ target_sources(
           ${CMAKE_CURRENT_SOURCE_DIR}/svd.cpp
           ${CMAKE_CURRENT_SOURCE_DIR}/inverse.cpp
           ${CMAKE_CURRENT_SOURCE_DIR}/cholesky.cpp
+          ${CMAKE_CURRENT_SOURCE_DIR}/utils.cpp
           ${CMAKE_CURRENT_BINARY_DIR}/compiled_preamble.cpp)
 
 if(IOS)