[SPARK-16184][SPARKR] conf API for SparkSession

## What changes were proposed in this pull request?

Add `conf` method to get Runtime Config from SparkSession

## How was this patch tested?

unit tests, manual tests

This is how it works in sparkR shell:
```
 SparkSession available as 'spark'.
> conf()
$hive.metastore.warehouse.dir
[1] "file:/opt/spark-2.0.0-bin-hadoop2.6/R/spark-warehouse"

$spark.app.id
[1] "local-1466749575523"

$spark.app.name
[1] "SparkR"

$spark.driver.host
[1] "10.0.2.1"

$spark.driver.port
[1] "45629"

$spark.executorEnv.LD_LIBRARY_PATH
[1] "$LD_LIBRARY_PATH:/usr/lib/R/lib:/usr/lib/x86_64-linux-gnu:/usr/lib/jvm/default-java/jre/lib/amd64/server"

$spark.executor.id
[1] "driver"

$spark.home
[1] "/opt/spark-2.0.0-bin-hadoop2.6"

$spark.master
[1] "local[*]"

$spark.sql.catalogImplementation
[1] "hive"

$spark.submit.deployMode
[1] "client"

> conf("spark.master")
$spark.master
[1] "local[*]"

```

Author: Felix Cheung <felixcheung_m@hotmail.com>

Closes #13885 from felixcheung/rconf.