Change the base class of lsst_noise_model, grid_selection, and spectroscopic_degrader

[ztq1996]

Problem & Solution Description (including issue #)

In this PR I have

1. added a general photometric error model class, which is an expansion of the old LSST error model class with the inclusion of the Roman and Euclid error model.
2. Change the base class of the grid_selection from degrader to selector, implement the _select function, and return a flag with the same length as that of the input data. I have make sure that the notebook in rail hub imports the new class and runs okay without the change needed.

Code Quality

• [ ] My code follows the code style of this project
• [ ] I have written unit tests or justified all instances of #pragma: no cover; in the case of a bugfix, a new test that breaks as a result of the bug has been added
• [ ] My code contains relevant comments and necessary documentation for future maintainers; the change is reflected in applicable demos/tutorials (with output cleared!) and added/updated docstrings use the NumPy docstring format
• [ ] Any breaking changes, described above, are accompanied by backwards compatibility and deprecation warnings

[hangqianjun]

Update on the selector (c.f. discussion in slack): resolving the problem with

data = input
some_criteria = something_about(data)
data2 = data[some_criteria]

another_criteria = something_else_about(data2)
data3 = data2[another_criteria]

return data3

where selection for data3 is untraceable directly to data. Option may be to add a tracker of the object ID or row index.

[ztq1996]

Great idea, I will implement the id approach tomorrow for the grid selection, because there are so many layers of selections

[jfcrenshaw]

LGTM!