Homodyne and heterodyne detectors not accepting modes in kwargs

[rachelchadwick]

Before posting a bug report

• I have searched exisisting GitHub issues to make sure the issue does not already exist.

Expected behavior

Homodyne(modes=[m],...) should apply a homodyne detector to mode m. Similarly for Heterodyne.

Actual behavior

It applies it to mode 0.

Reproduces how often

Homodyne(modes=[m],...) always applies it to mode 0.
Note that Homodyne(...)[m] works correctly, applying it to mode m.
Similarly for Heterodyne.

System information

Python version:            3.8.0
Platform info:             Windows-10-10.0.19041-SP0
Installation path:         C:\Users\Rache\AppData\Local\Programs\Python\Python38\lib\site-packages\mrmustard
Mr Mustard version:        0.1.1
Numpy version:             1.19.2
Numba version:             0.53.1
Scipy version:             1.4.1
The Walrus version:        0.17.0
TensorFlow version:        2.7.0
Torch version:             None

Source code

from mrmustard.lab import *
import numpy as np
from mrmustard.utils.graphics import mikkel_plot
import matplotlib.pyplot as plt

# Initial state is two modes with "diagonal" (angle=pi/2) squeezed state in mode 0
# and "vertical" (angle=0) squeezed state in mode 1
S1 = Sgate(modes=[0], r=1, phi=np.pi/2)
S2 = Sgate(modes=[1], r=1, phi=0)
initial_state = Vacuum(2) >> S1 >> S2

# Because the modes are separable, measuring in one mode should leave
# the state as simply the state in the unmeasured mode unchanged

# A homodyne measurement on the second mode should leave
# the diagonal state in the first mode but doesn't
final_state = initial_state << Homodyne(modes=[1],quadrature_angle=0,result=[0.3])
mikkel_plot(np.array(final_state.dm()))

# But coded like this does work
final_state = initial_state << Homodyne(quadrature_angle=0,result=[0.3])[1]
mikkel_plot(np.array(final_state.dm()))

plt.show()

Tracebacks

No response

Additional information

Proposed fix from @ilan-tz:

`Homodyne` initializes `DisplacedSqueezed` which initializes `State`.
Line 370 in states.py currently is 
`State.__init__(self, cov=cov, means=means, cutoffs=cutoffs)` 
but should be
`State.__init__(self, cov=cov, means=means, cutoffs=cutoffs, modes=modes)`

Similarly for `Heterodyne`, line 112 in `Coherent` should have the modes kwarg added.

In fact, in general the modes of the states do not seem to depend on the kwarg `modes` but just the length of the list of parameters given.