[Datasets] Do not eagerly execute first block for read_xxx API

[c21]

Signed-off-by: Cheng Su scnju13@gmail.com

Why are these changes needed?

This PR is the followup of #31286 (review). The change includes:

• read_api.py:read_datasource(): Remove the logic to eagerly execute first block for read.
• Dataset.schema(): Change default value of fetch_if_missing from False to True. So always trigger execution if schame is missing.
• ExecutionPlan.schema(): if plan is having lazy block list as output, execute the first block only to get schema, instead of executing all blocks.
• Other files: unit test change to work with new logic.

Related issue number

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 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 :(

[c21]

The failed book-documentation looks unrelated to this PR. Seeing it also has same failure on other PR.

[clarkzinzow]

LGTM!