[Core] Add cached memory to available memory

[lixin-wei]

Why are these changes needed?

Cached memory should also be considered as available memory.

When I start ray cluster in a machine that has low available mem but high buff/cache memory, a rayOutOfMemoryError is raised.

2020-08-10 16:26:03,954 ERROR worker.py:1085 -- Possible unhandled error from worker: ray::JobWorker.__init__() (pid=42130, ip=11.166.237.22)
  File "python/ray/_raylet.pyx", line 435, in ray._raylet.execute_task
  File "/home/admin/ray/python/ray/memory_monitor.py", line 128, in raise_if_low_memory
    self.error_threshold))
ray.memory_monitor.RayOutOfMemoryError: More than 95% of the memory on node corgiunit-eu95-5.rz00b.stable.alipay.net is used (15.59 / 16.0 GB). 

$ free -h
              total        used        free      shared  buff/cache   available
Mem:            16G        5.9G        980M        3.6G        9.1G        980M
Swap:            0B          0B          0B

Related issue number

Checks

• 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/latest/.
• I've made sure the tests are passing. Note that there might be a few flaky tests, see the recent failure rates at https://ray-travis-tracker.herokuapp.com/.
• Testing Strategy
  • Unit tests
  • Release tests
  • This PR is not tested (please justify below)

[AmplabJenkins]

Can one of the admins verify this patch?

[AmplabJenkins]

Test FAILed.
Refer to this link for build results (access rights to CI server needed):
https://amplab.cs.berkeley.edu/jenkins//job/Ray-PRB/29656/
Test FAILed.

[suquark]

.available/(1024**3)

[lixin-wei]

.available/(1024**3)

My mistake, fixed, thanks a lot.

[lixin-wei]

@suquark Are there any other problems? Could you please approve this PR so that it can be merged?

fixed