CL util function cleanup

[j-c-c]

Documenting some cleanup tasks to follow the Common-lines reorg work. Adding the associated code review comments below for reference (see #1341 for more info):

I have 3 suggestions about the same sort of code. Its not new code, just migrated. Maybe it can be improved (maybe not). Make a followup issue to review those in the future after the reorg stuff. Low priority, just something I might have missed before. Thanks

In src/aspire/abinitio/commonline_utils.py:

>  
-                Rj_tilde = Ri_tildes[j]
+        # Generate shifted versions of images.
+        pf_i_shifted = np.array([pf_i * shift_phase for shift_phase in shift_phases])

can we just multiply these arrays....

In src/aspire/abinitio/commonline_utils.py:

>  
-                # Perform correlation, corrs is shape n_shifts x len(theta_ijs).
-                corrs = np.array(
-                    [
-                        np.dot(pf_i_shift[c1], np.conj(pf_j[c2]))
-                        for pf_i_shift in pf_i_shifted
-                        for c1, c2 in zip(c1s, c2s)
-                    ]
-                )
+            # Compute all possible rotations between the i'th and j'th images.
+            Us = np.array(

Similarly here, seems like a broadcast would work? This might be a rare case where einsum could be useful.

In src/aspire/abinitio/commonline_utils.py:

>  
-                # Reshape to group by shift and symmetric order.
-                corrs = corrs.reshape((n_shifts, order, len(theta_ijs) // order))
+            # Find the angle between common lines induced by the rotations.
+            c1s = np.array([[-U[1, 2], U[0, 2]] for U in Us])
+            c2s = np.array([[U[2, 1], -U[2, 0]] for U in Us])
+
+            # Convert from angles to indices.
+            c1s = _cl_angles_to_ind(c1s, n_theta)
+            c2s = _cl_angles_to_ind(c2s, n_theta)
+
+            # Perform correlation, corrs is shape n_shifts x len(theta_ijs).
+            corrs = np.array(

This would probably be clearer without the nested list comprehension. And you might not need to reshape it ....