Adding Deterministic SNR

[ark0015]

Here are some test plots to show it works. The figures are an addition to the GWBSensitivity tutorial plots.

[Hazboun6]

@ark0015 This is looking pretty good! Really good start. I think it just needs a few tweaks. First let's refer to Issue #1 which is what this pull request is concerning.

Since the DeterSensitivityCurve works for all deterministic curves, I think I'd rather not assume that it calculates the SNR for a circular binary. Take a look at the SNR for the GWBSensitivityCurve it doesn't actually assume the form of the background.

One other thing is that the proper use of the astropy.units would be to call return snr.to('').value, so that it converts to the correct units for you .

[ark0015]

So I tried a few different things, but I think just generalizing the naming to just h_0 is what I settled on. It produces identical results to those in my first commits.

Originally, I thought it would be a good idea to just move the SNR calculation in GWBSensitivityCurve out of the class to use as a general function that takes in S_eff, Sh, freqs, Tspan for any sensitivity curve and source Sh. Then the Sh for a monochromatic source would be eqn. 50 from the paper. It definitely is the most general case, and would work for any strain PSD.
The issue was that this method returned a much higher SNR estimation than the one given in my examples above for eqn. 76 for the monochromatic case using eqn. 50. I suspect because the delta approximations spread the power over so many frequencies that it really is not a good approximation of the delta function over the observation time at such low frequencies. I can provide plots that show this if you would like to see.