Support multiple files random and sequential read in StressWorkerBench

[bobbai00]

What changes are proposed in this pull request?

1. add several additional parameters in WorkerBenchParameters
2. support multiple files random read&sequential read benchmark in StressWorkerBench

Why are the changes needed?

In order to comparing the worker's read performance more comprehensively, I switch the single test file to multiple test files, and add two modes of reading these files:

• sequential read: each thread will try to read the entire file of every single test file path. When doing sequential read, --random-read shoud NOT be set.
• random read: in the prepare stage, several random offsets and lengths will be generated and recorded. And each thread will try to read each test file according to the offset and length. When doing random read, --random-read, --random-seed, --random-max-length and --random-min-length should be set.

Does this PR introduce any user facing changes?

Nope.

[alluxio-bot]

Automated checks report:

• Commits associated with Github account: PASS
• PR title follows the conventions: FAIL
  • The title of the PR does not pass all the checks. Please fix the following issues:
    • Title is too long (78 characters). Must be at most 72 characters.

Some checks failed. Please fix the reported issues and reply 'alluxio-bot, check this please' to re-run checks.

[alluxio-bot]

Automated checks report:

• Commits associated with Github account: PASS
• PR title follows the conventions: PASS

All checks passed!

[beinan]

one test failing is not related, this pr did not change any function code

[beinan]

alluxio-bot, merge this please.

[alluxio-bot]

Check evaluation failed: Github action modules: (alluxio.server.ft.journal.raft.,alluxio.server.ft.journal.ufs.) concluded as failure, which is not success or skipped

[jiacheliu3]

alluxio-bot, merge this please