[data] Handle errors in SplitCoordinator when generating a new epoch

[raulchen]

Why are these changes needed?

Previously if an error happens while launching the executor while generating a new epoch, we don't handle the error properly, thus users will get a confusing error. And the actual root cause will be hidden.

E                         File "/Users/chenh/code/ray/python/ray/data/_internal/iterator/stream_split_iterator.py", line 197, in start_epoch
E                           epoch_id = self._barrier(split_idx)
E                         File "/Users/chenh/code/ray/python/ray/data/_internal/iterator/stream_split_iterator.py", line 285, in _barrier
E                           assert self._output_iterator is not None
E                       AssertionError

This PR solves this issue by storing the error and re-raise it from all threads.

Related issue number

Closes #45225

Checks

• I've signed off every commit(by using the -s flag, i.e., git commit -s) in this PR.
• I've run scripts/format.sh to lint the changes in this PR.
• I've included any doc changes needed for https://docs.ray.io/en/master/.
  • I've added any new APIs to the API Reference. For example, if I added a
    method in Tune, I've added it in doc/source/tune/api/ under the
    corresponding .rst file.
• I've made sure the tests are passing. Note that there might be a few flaky tests, see the recent failures at https://flakey-tests.ray.io/
• Testing Strategy
  • Unit tests
  • Release tests
  • This PR is not tested :(