[Data] Skip schema call in to_tf if tf.TypeSpec is provided

[c21]

Why are these changes needed?

This PR is to skip Dataset.schema() call from Dataset.to_tf(). Dataset.schema() relies on limit to early stop execution, and sometimes the stop is not triggered timely so a lot of tasks get executed. This introduced problem to cause memory spilling. In addition, sometimes, it returns None (does not work with limit push down), and it breaks followed logic in to_tf, which all relies on schema() to work.

In this PR:

• Introduce two optional parameters in to_tf: feature_type_spec and label_type_spec (by default they are None). So user can set tf.TypeSpec explicitly and the Dataset.schema() call will be skipped.

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 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 :(