Introduce JuiceFS Configuration

[MarcusSorealheis]

Description

There are limitations to a distributed system where a cluster of workers are independently working through the same queue of tasks. This configuration will enable workers in a cluster to have immediate visibility into updates to the CAS. Worker 1 can upload digest 59859858y5, and Workers 2-100 can immediately read it immediately.

There have been some interesting race conditions and other related issues with the filesystem that have come up recently (#1863, #1928, #1931, #1934, etc.). The list is much longer in reality because this was always an issue.

Fixes #1949

Type of change

Please delete options that aren't relevant.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to
  not work as expected)
• This change requires a documentation update

How Has This Been Tested?

Please also list any relevant details for your test configuration

Checklist

• Updated documentation if needed
• Tests added/amended
• bazel test //... passes locally
• PR is contained in a single commit, using git amend see some docs

---

This change is 

[palfrey]

This doesn't work without docker plugin install juicedata/juicefs being run first

[palfrey]

I can't find the linked file

[palfrey]

Not seeing the juicefs with k8s doc

[palfrey]

This doesn't work without docker plugin install juicedata/juicefs being run first

Actually, more problems. juicefs-mount service gives me Error response from daemon: Encountered remote "application/vnd.docker.plugin.v1+json"(plugin) when fetching