Fix for "Reader argument to ascii.read was deprecated" warning

[javierggt]

Description

This fixes the Reader argument to ascii.read was deprecated warning.

Fixes #164

Interface impacts

Testing

Unit tests

• HEAD
• Mac

Functional tests

I specifically ran pytest agasc/tests/test_agasc_2.py::test_against_ds_agasc on HEAD, both on the branch and on master, to verify that the master version gives the warning and the reader-deprecated branch does not.

[taldcroft]

@javierggt - testing needs to be done on HEAD linux since it depends on the ASCDS mp_get_agasc tool running. Also the linting CI is failing.

[javierggt]

@taldcroft I did notice the linting CI is failing, but it is failing in a file I did not touch in this PR. Not only that, the file is in the dev directory. And on top of that, the linting is not reporting all the failures I saw offline when I ran black, after checking the details of the workflow.

Still, I can fix that if you want. I just thought this was a trivial PR.

[javierggt]

@taldcroft, I fixed the linting issue.

I do not think it is reasonable in general to require fixing linting issues in code unrelated to the PRs. In this case it is simple enough, but I'm guessing the failure was related to a newer black version since that file's last commit.

[jeanconn]

Yeah, it has annoyed me in the past -- I thought github now had some hooks to try to run formatting checks only on the areas that have changed for the PR but it looked to me like it that wasn't worth it. I'm not sure if the way forward is to just fix formatting issues in the unrelated PR as you've done.

[taldcroft]

@jeanconn - there exist kludgy ways to lint only changed code, but we don't (and won't) do that.

@javierggt - I agree with the principle that your PR is not responsible for cleaning linting problems. But in practice it needs to be fixed somewhere. The other option is to fix it in another PR and then rebase your PR on the fixed one. Your choice but I think just fixing it one PR is a little easier.

BTW, I did not look at the CI report but rather just noted that as a blocker before I would approve. I'd like to stick with our CI rules though and not merge anything that doesn't pass.

[taldcroft]

LGTM.