[Executorch][llm] Make mask tensor float only for sdpa

Now that we support quantized sdpa query tensor can be quantized and attention mask can be float (the only type allowed).
So this check doesnt make sense anymore.

Differential Revision: [D77516821](https://our.internmc.facebook.com/intern/diff/D77516821/)

ghstack-source-id: 293661338
Pull Request resolved: #12131

Author: kimishpatel

extension/llm/custom_ops/op_sdpa.cpp

@@ -59,8 +59,8 @@ bool validate_flash_attention_args(
 
   ET_CHECK_OR_RETURN_FALSE(
       !attn_mask.has_value() ||
-          attn_mask.value().scalar_type() == query.scalar_type(),
-      "Attention mask must be a 2D tensor");
+          attn_mask.value().scalar_type() == ScalarType::Float,
+      "Attention mask must be a Float tensor");
 
   ET_CHECK_OR_RETURN_FALSE(
       is_contiguous_dim_order(query.dim_order().data(), query.dim()),