[GraphBolt] CachePolicy Writer lock for read_async correctness.

[mfbalin]

Description

In async feature loading pipeline, after we insert into the policy, values may not be still written into the cache yet. Using writer locks in addition to reader locks, we can act as if inserted values whose payload has not been inserted yet do not exist during our reads.

The PR is a bit large because I needed to do the change for all 4 cache policies at once.

Checklist

Please feel free to remove inapplicable items for your PR.

• The PR title starts with [$CATEGORY] (such as [NN], [Model], [Doc], [Feature]])
• I've leverage the tools to beautify the python and c++ code.
• The PR is complete and small, read the Google eng practice (CL equals to PR) to understand more about small PR. In DGL, we consider PRs with less than 200 lines of core code change are small (example, test and documentation could be exempted).
• All changes have test coverage
• Code is well-documented
• To the best of my knowledge, examples are either not affected by this change, or have been fixed to be compatible with this change
• Related issue is referred in this PR
• If the PR is for a new model/paper, I've updated the example index here.

Changes

[dgl-bot]

To trigger regression tests:

• @dgl-bot run [instance-type] [which tests] [compare-with-branch];
  For example: @dgl-bot run g4dn.4xlarge all dmlc/master or @dgl-bot run c5.9xlarge kernel,api dmlc/master

[dgl-bot]

Commit ID: df2fe6d

Build ID: 1

Status: ✅ CI test succeeded.

Report path: link

Full logs path: link

[frozenbugs]

LGTM