docs: recommend using guppy.load_pytket for pure unitary circuits

sphinx/migration_guide.md

@@ -46,9 +46,14 @@ qft_func = guppy.load_pytket("qft_func", qft2_circ)
 
 We now have a Guppy function called `qft_func` which we can compile or use as a subroutine in other functions.
 
-Note that by default [guppy.load_pytket][load_pytket_docs] will create Guppy functions which use arrays of qubits as inputs. This means that our `qft_func` above will have take an array of two qubits as input. 
+By default [guppy.load_pytket][load_pytket_docs] will create Guppy functions which use arrays of qubits as inputs. This means that our `qft_func` above will take an array of two qubits as input. 
 If we want the function to take two separate qubit arguments, we can specify `use_arrays=False` in [guppy.load_pytket][load_pytket_docs]. Also note that by default, circuits with separate quantum registers become Guppy functions that take multiple arrays of qubits as input. 
 
+Using [guppy.load_pytket][load_pytket_docs] works best when the loaded pytket [Circuit](inv:pytket#circuit_class) contains only unitary quantum gates. This is because Guppy allows
+ qubits to be explicitly allocated at runtime and deallocated by measurement. By contrast, pytket
+  treats qubits as alive for the entirety of the quantum program.
+   It is therefore best practice to use [guppy.load_pytket][load_pytket_docs] for unitary subroutines,
+    which map directly to Guppy functions which borrow qubits. Measurement and classical logic should ideally be done directly in Guppy. 
 
 ### How to deal with operations unsupported by Guppy