FIX: handle empty first Eyelink recording block

[varunkasyap]

Reference issue (if any)

Fixes #13550.

What does this implement/fix?

Fixes a crash when reading Eyelink ASCII files whose first recording block contains no sample data. The reader now handles this edge case and includes a test.

Additional information

Test suite passed in my local

[varunkasyap]

Hey, I couldn’t understand the reason for the failing checks. Can someone please explain?

[varunkasyap]

Hey, just wanted to check why the welcome bot didn’t get triggered for my PR?

[scott-huberty]

Thanks @varunkasyap ! I will take a look this week.

[scott-huberty]

Hey @

Hey, I couldn’t understand the reason for the failing checks. Can someone please explain?

Short answer is that you need to create a CircleCI account. if you sign in to CircleCI with your GitHub account, then the CI will be able to confirm that there is a CircleCI account associated with your GitHub username, and the check should run.

in MNE-Python, we use Github Actions to run some of our CI's, and we use CircleCI to run some CI's. For security reasons, we prevent PRs from users who dont have CircleCI accounts from triggering CircleCI hosted tests.

[varunkasyap]

Short answer is that you need to create a CircleCI account. if you sign in to CircleCI with your GitHub account, then the CI will be able to confirm that there is a CircleCI account associated with your GitHub username, and the check should run.

Thanks @scott-huberty

I think this PR is now ready for review. Do I need to mention this change somewhere in the docs?