### What changes were proposed in this pull request?
This PR proposes to restructure User Guide in PySpark documentation for pandas APIs on Spark.
**Before**
![Screen Shot 2021-06-08 at 8 47 41 PM](https://user-images.githubusercontent.com/6477701/121179493-cb85e280-c89a-11eb-8b93-552ebe7cd0a8.png)
**After**
![Screen Shot 2021-06-08 at 8 46 58 PM](https://user-images.githubusercontent.com/6477701/121179419-b3ae5e80-c89a-11eb-82a0-6dabbf1de12d.png)
Note that I mostly just moved the contents around except minor changes:
- Removing some questions in FAQ that don't make sense in Apache Spark
- Rename a subtitle "Working with pandas and PySpark" to "From/to pandas and PySpark DataFrames"
For renaming Koalas to either pandas-on-Spark or pandas APIs on Spark, it will be done at SPARK-35591
### Why are the changes needed?
For better readability.
### Does this PR introduce _any_ user-facing change?
Yes, it restructures the documentation as shown above.
### How was this patch tested?
I manually built the docs and tested.
Closes#32820 from HyukjinKwon/SPARK-35647.
Authored-by: Hyukjin Kwon <gurwls223@apache.org>
Signed-off-by: Hyukjin Kwon <gurwls223@apache.org>