Dry/wet (i.e. include partial raw signal) parameter on IIR notch filter before convolution with other filters

Question

I have a group of scipy.signal.iirnotch filters convolving with one another and I would like each to implement independent dry/wet values for each of them. I know I can achieve a pseudo-dry/wet parameter by adding them to one another instead of performing a convolution on them but that is janky and difficult to control.

I am just at the start of my DSP learning but I haven't been able find any documentation/information on something like this so I have no idea where else to start other than poking around at the coefficients (to no avail). Here is the isolated relevant code so far:

from scipy import signal
notch_freqs = (220, 440, 880) # could be any frequencies, these are just placeholders
notch_coeffs = [signal.iirnotch(freq, BANDWIDTH, samplerate) for freq in notch_freqs]

# here would be the step in applying the dry/wet to each coefficient in notch_coeffs

# and then the convolution would happen...
def convolve_notches():  # I know this can probably be optimized but it works for now
    filter_conv_a = signal.fftconvolve(*[filt[0] for filt in notch_coeffs[:2]])
    filter_conv_b = signal.fftconvolve(*[filt[1] for filt in notch_coeffs[:2]])

    if len(notch_coeffs) > 2:
         for filt in notch_coeffs[:2]:
            filter_conv_a = signal.fftconvolve(filter_conv_a, filt[0])
            filter_conv_b = signal.fftconvolve(filter_conv_b, filt[1])
    return filter_conv_a, filter_conv_b

# ... then apply using signal.filtfilt

This is not a real-time audio application so I'm not too concerned about timing. Rereading this, I want to clarify dry/wet; here's an ASCII representation:

Wet: 25%       50%         100%
---\    /----\     /----\       /----
    \__/      \   /      \     /
               \_/        \   /
                           | |
                           | |