Moved flash_attn assert close to the caller

Moved the flash_attn package availability assert close to the caller _apply_dense_attention. This allows the non cuda devices run the model.

modeling_phi3_small.py

@@ -418,6 +418,8 @@ class Phi3SmallSelfAttention(nn.Module):
             avoid doing that.
 
         """
+        assert is_flash_attention_available, "Flash Attention is not available, but is needed for dense attention"
+        
         attention_dropout_prob = self.attention_dropout_rate if self.training else 0.0
         # Get into the correct shape for the Flash Attention API
         # shape: (bs, seq_len, nqp, hn)