More accelerate optimizations (#427)

* Add more tracing to the whisper example.

* Support accelerate in more examples.

* Use accelerate for pointwise functions.

* Use accelerate for binary operations too.

* Bugfix for binary operation: use the rhs before the lhs.

candle-examples/examples/whisper/main.rs

@@ -1,9 +1,11 @@
 // https://github.com/openai/whisper/blob/main/whisper/model.py/rgs
 // TODO:
-// - kv-cache support?
 // - Batch size greater than 1.
 // - More token filters (SuppressBlanks, ApplyTimestampRules).
 
+#[cfg(feature = "accelerate")]
+extern crate accelerate_src;
+
 #[cfg(feature = "mkl")]
 extern crate intel_mkl_src;