Data Argument for creating parameters

[Simardeep27]

This argument allows the user to create parameters based on scipy.fit() method. The user won't be required to input the required arguments for the selected distributions and simply pass the data for which they want to create a parameter object.

Syntax:

eq.Parameter(distribution='gaussian',data=arr)


Currently, data parameter is added for:

1. Uniform
2. Gaussian
3. Lognormal
4. Exponential
5. Beta

The required documentation has been added and test cases have been added in tests/test_parameter.py

[psesh]

One concern is that only four distributions (uniform, exponential log-normal, and normal) are tested in test_parameters.py, but all distributions seem to be modified with this change. Additionally, it would be good to see a test of how this propgates into estimates of the output moments for a function. I'd be happy to add the needful to test_statistics.py.