Merge branch 'main' into py_refactor

docs/sphinx/specification/cudaq/operations.rst

@@ -99,4 +99,8 @@ qubit state is a :code:`|0>` (positive polarity) or :code:`|1>` (negative polari
 By default all control qubits are assumed to convey positive polarity. 
 The syntax for negating the polarity is the not operator preceeding the
 control qubit (e.g., :code:`x<cudaq::ctrl>(!q, r)`, 
-for :code:`cudaq::qubits` :code:`q` and :code:`r`).
+for :code:`cudaq::qubits` :code:`q` and :code:`r`). Negating the polarity of
+control qubits is supported in :code:`swap` and the gates in sets :code:`NAME`
+or :code:`ROTATION_NAME`. The negate notation is only supported on control
+qubits and not target qubits. So negating either of the target qubits in the
+:code:`swap` operation is not allowed.

docs/sphinx/specification/cudaq/synthesis.rst

@@ -54,6 +54,14 @@ The :code:`cudaq::control(...)` function takes as input an instantiated pure
 device quantum kernel, an std::range of control qubits (:code:`cudaq::qreg`
 or :code:`cudaq::qspan`), and the remaining arguments for the kernel itself.
 Compiler implementations are free to synthesize multi-controlled operations
-using any pertinent synthesis strategy available. The :code:`cudaq::adjoint(...)`
-function takes as input an instantiated pure device quantum kernel
-(or specified template type) and the remaining arguments for the kernel. 
+using any pertinent synthesis strategy available. Qubits may be aggregated into
+a range of control qubits with or without the use of the :code:`operator!`
+`negated polarity operator <https://nvidia.github.io/cuda-quantum/specification/cudaq/operations.html>`_.
+
+.. code-block:: cpp
+
+   cudaq::control(kernel{}, {qubit0, !qubit1}, kernel_arg);
+
+The :code:`cudaq::adjoint(...)` function takes as input an
+instantiated pure device quantum kernel (or specified template type)
+and the remaining arguments for the kernel.