Add the cuda dequantize f16 kernels. (#2137)

* Add the cuda dequantize f16 kernels.

* Expose the cuda kernels.

* Add some testing + fix.

* Test the other cases too.

* A few more tests.

* Add an environment variable to enable the dequantize f16 + matmul behavior.

candle-core/src/quantized/dummy_cuda.rs

@@ -24,6 +24,10 @@ impl QCudaStorage {
         Err(Error::NotCompiledWithCudaSupport)
     }
 
+    pub fn dequantize_f16(&self, _elem_count: usize) -> Result<CudaStorage> {
+        Err(Error::NotCompiledWithCudaSupport)
+    }
+
     pub fn quantize(&mut self, _src: &CudaStorage) -> Result<()> {
         Err(Error::NotCompiledWithCudaSupport)
     }