Fixes #6942

Description

This might look like a large PR but most changes are mechanical i.e. updates to import statements and guava dependency in various pom files.

Here is the meaty part...

1. A new module "shaded-guava" is introduced that produces classes in "org.apache.druid.com.google.common" package by shading real guava. See shaded-guava/pom.xml .
2. pom.xml in all modules are made to depend upon shaded guava jar above instead of real guava, that change looks like

-      <groupId>com.google.guava</groupId>
+      <groupId>org.apache.druid.com.google.guava</groupId>

3. In all Java files, import com.google.common.X is replaced by import org.apache.druid.com.google.guava.X
4. "Organize Imports" is done to all Java files to correct import ordering or checkstyles fails
5. Some imports in cloudfiles-extensions and sql modules are reverted to use import com.google.common.X because their dependencies jclouds-core and calcite respectively mandate using real import com.google.common.X in various methods exposed by those libraries. But this is minimal.
6. some imports that used org.apache.curator.shaded.com.google.common.X , i.e. shaded version of guava used by curator, continue to use same. that is independent.

(this approach was briefly described in #6942 (comment) )

End Result is that Druid source code is free from real guava and can use any version of guava inside shaded-guava module, few exceptions in (5) as described above.
we still ship guava-0.16.1.jar like before, so no external integrations should see any impact. User should be able to switch this with their favorite version of guava jar at runtime and Druid source code be fine because it uses druid-shaded-guava-0.x.y.jar

Existing Unit, Integration Tests and Travis builds should cover all of the changes except for Hadoop Integration that I could not verify due to not having quick access to such setup. However, given the way this patch works and that we continue to ship existing guava jar, Hadoop Integration should see no change.

#10965 should also be done after this PR is accepted.

This PR has:

• been self-reviewed.
  • using the concurrency checklist (Remove this item if the PR doesn't have any relation to concurrency.)
• added documentation for new or modified features or behaviors.
• added Javadocs for most classes and all non-trivial methods. Linked related entities via Javadoc links.
• added or updated version, license, or notice information in licenses.yaml
• added comments explaining the "why" and the intent of the code wherever would not be obvious for an unfamiliar reader.
• added unit tests or modified existing tests to cover new code paths, ensuring the threshold for code coverage is met.
• added integration tests.
• been tested in a test Druid cluster.