[Java] Simplify Java worker configuration (#2938)

## What do these changes do?
Previously, Java worker configuration is complicated, because it requires setting environment variables as well as command-line arguments.

This PR aims to simplify Java worker's configuration. 
1) Configuration management is now migrated to [lightbend config](https://github.com/lightbend/config), thus doesn't require setting environment variables.
2) Many unused config items are removed.
3) Provide a simple `example.conf` file, so users can get started quickly.
4) All possible options and their default values are declared and documented in `ray.default.conf` file.

This PR also simplifies and refines the following code:
1) The process of `Ray.init()`.
2) `RunManager`.
3) `WorkerContext`. 

### How to use this configuration?
1. Copy `example.conf` into your classpath and rename it to `ray.conf`.
2. Modify/add your configuration items. The all items are declared in `ray.default.conf`.
3. You can also set the items in java system prosperities.

Note: configuration is read in this priority:
System properties > `ray.conf` > `ray.default.conf`

## Related issue number
N/A

java/runtime/src/main/resources/log4j.properties

@@ -1,20 +1,17 @@
-# define default properties here
-logging.level=WARN
-logging.path=./run/logs
-logging.file.name=core
-logging.max.log.file.num=10
-logging.max.log.file.size=500MB
+ray.logging.level=INFO
 
-log4j.rootLogger=${logging.level}, stdout, core
+ray.logging.stdout=org.apache.log4j.ConsoleAppender
+ray.logging.file=org.apache.log4j.varia.NullAppender
 
-log4j.appender.stdout=org.apache.log4j.ConsoleAppender
+log4j.rootLogger=${ray.logging.level}, stdout, file
+
+log4j.appender.stdout=${ray.logging.stdout}
 log4j.appender.stdout.layout=org.apache.log4j.PatternLayout
 log4j.appender.stdout.layout.ConversionPattern=%d{yyyy-MM-dd HH:mm:ss} %p %c{1} [%t]: %m%n
 
-log4j.appender.core=org.apache.log4j.RollingFileAppender
-log4j.appender.core.File=${logging.path}/${logging.file.name}.log
-log4j.appender.core.Append=true
-log4j.appender.core.MaxFileSize=${logging.max.log.file.size}
-log4j.appender.core.MaxBackupIndex=${logging.max.log.file.num}
-log4j.appender.core.layout=org.apache.log4j.PatternLayout
-log4j.appender.core.layout.ConversionPattern=%d{yyyy-MM-dd HH:mm:ss} %p %c{1} [%t]: %m%n
+# Set the file appender to null by default. If `ray.redirect-output` config is set to true,
+# this appender will be set to a real file appender.
+log4j.appender.file=${ray.logging.file}
+log4j.appender.file.File=${ray.logging.file.path}
+log4j.appender.file.layout=org.apache.log4j.PatternLayout
+log4j.appender.file.layout.ConversionPattern=%d{yyyy-MM-dd HH:mm:ss} %p %c{1} [%t]: %m%n

java/runtime/src/main/resources/ray.default.conf

@@ -0,0 +1,79 @@
+// This file contains default values of all Ray configurations.
+// Users should define their own 'ray.conf' file in the classpath,
+// or use Java properties, to overwrite these values.
+
+ray {
+  // ----------------------
+  // Basic configurations
+  // ----------------------
+
+  // This is the path to the directory where Ray is installed, e.g.,
+  // something like /home/ubmutu/ray. This can be an absolute path or
+  // a relative path from the current working directory.
+  home: ""
+
+  // IP of this node. if not provided, IP will be automatically detected.
+  node-ip: ""
+
+  // Run mode, available options are:
+  //
+  // `SINGLE_PROCESS`: Ray is running in one single Java process, without Raylet backend,
+  //                   object store, and GCS. It's useful for debug.
+  // `CLUSTER`: Ray is running on one or more nodes, with multiple processes.
+  run-mode: CLUSTER
+
+  // Available resources on this node, for example "CPU:4,GPU:0".
+  resources: ""
+
+  // If worker.mode is DRIVER, specify the driver id.
+  // If not provided, a random id will be used.
+  driver.id: ""
+
+  // Root dir of log files.
+  log-dir: /tmp/ray/logs
+
+  // If true, output of worker processes will be redirected to log files.
+  // Otherwise, output will be printed to console.
+  redirect-output: true
+
+  // Custom `java.library.path`
+  // Note, do not use `dir1:dir2` format, put each dir as a list item.
+  library.path: []
+
+  // Custom classpath.
+  // Note, do not use `dir1:dir2` format, put each dir as a list item.
+  classpath = []
+
+  // ----------------------
+  // Redis configurations
+  // ----------------------
+  redis {
+    // The address of the redis server to connect, in format `ip:port`.
+    // If not provided, Ray processes will be started locally, including
+    // Redis server, Raylet and object store.
+    address: ""
+    // If `redis.server` isn't provided, which port we should use to start redis server.
+    head-port: 6379
+    // If `redis.server` isn't provided, how many Redis shards we should start in addition to the
+    // primary Redis shard. The ports of these shards will be `head-port + 1`, `head-port + 2`, etc.
+    shard-number: 1
+  }
+
+  // ----------------------------
+  // Object store configurations
+  // ----------------------------
+  object-store {
+    // RPC socket name of object store
+    socket-name: /tmp/ray/sockets/object_store
+    // Initial size of the object store.
+    size: 10 MB
+  }
+
+  // ----------------------------
+  // Raylet configurations
+  // ----------------------------
+  raylet {
+    // RPC socket name of Raylet
+    socket-name: /tmp/ray/sockets/raylet
+  }
+}