Clarify placeholder IR that delegates need to support in ExecuTorch

📚 The doc issue

We do not specify the IR for placeholders/graph inputs that delegates need to support.

For example, today I think placeholders will usually be tensors, with floats and ints inputs being lifted to rank0 tensors. But raw symints also show up as placeholders in the presence of graph breaks. For example,

Lowered module 0:

def forward(...):
    sym_size: "Sym(s53)" = torch.ops.aten.sym_size.int(attention_mask, 1);  attention_mask = None
    return [sym_size, aten_expand_copy_default, aten_add_tensor_1]

Lowered module 1:

def forward(aten_embedding_default_2: "f32[1, s53, 768]", sym_size: "Sym(s53)"):
    aten_view_copy_default_3: "f32[1, s53, 768]" = executorch_exir_dialects_edge__ops_aten_view_copy_default(aten_permute_copy_default_3, [1, sym_size, 768]); 

We should clarify placeholder IR that delegates are expected to support.

Suggest a potential alternative/fix

No response

cc @JacobSzwejbka @angelayi

Start with the placeholder and lowered-module examples in this issue, then trace where ExecuTorch documents delegate input requirements. Document which placeholder forms delegates must support, including tensor inputs and raw SymInts, with examples that make the expected IR unambiguous.