Changes to fit coefficient outputs

Nobody has claimed this yet.

1. Can we reverse the order of the polynomial that is put out? Currently it lists Ax^4 + Bx^3 + Cx^2 + Dx + F. This always seems backward to us. Can we instead see F + Dx + Cx^2 + Bx^3 + Ax^4?

2. Can we just see a number, rather than the value written as an exponent? For example, we would prefer 0.0678 rather than 6.78E-2. (For small numbers. Obviously for 3e100 this isn't going to work. We may need an if to deal with this. Python naturally does this (see here https://datagy.io/python-scientific-notation/) but not when formatting. And we need formatting to force item 3 I think.

3. Have Curvallis give a fit with more decimal points? At least 8.

Thanks,

Start by locating the code path that formats and displays polynomial fit coefficients. Update the output to reverse coefficient order, use fixed-point notation where practical, and show at least eight decimal places. Verify the displayed results with both small and very large coefficient values.

Written by the indexing model from the issue text.