COS filesystem high availability support (#43)

* Add gradle tasks to minimal and dependencies to maven local

* Add capability to dagger to read python udfs from Ali(oss) and Tencent(cosn) storage services
Given the configuration provided correctly. Set the below environment
variables accordingly to access the files stored in the respective
bucket.

Ali(oss)
- OSS_ACCESS_KEY_ID
- OSS_ACCESS_KEY_SECRET

Tencent(cos)
- COS_SECRET_ID
- COS_SECRET_KEY
- COS_REGION

* OSS client endpoint should be configurable via ENV variable

* COS filesystem high availability support
If you need to use COS filesystem for the dagger, provide the cos
bucket/key configuration in the state.backend.fs.checkpointdir,
state.savepoints.dir, high-availability.storageDir to flinkdeployment
manifest.

If the filesystem protocol begins with cosn for the above
configurations, dagger uses the below configurations provided in the
flinkdeployment manifest file.

    fs.cosn.impl: org.apache.hadoop.fs.CosFileSystem
    fs.AbstractFileSystem.cosn.impl: org.apache.hadoop.fs.CosN
    fs.cosn.userinfo.secretId: <secretID>
    fs.cosn.userinfo.secretKey: <secretKey>
    fs.cosn.bucket.region: <region>
    fs.cosn.bucket.endpoint_suffix: <tencent-provided-prefix.xyz.com>

* Fix checkstyle and made constants as static variables

---------

Co-authored-by: Raju G T <[email protected]>
Co-authored-by: rajuGT <[email protected]>

dagger-common/build.gradle

@@ -65,6 +65,7 @@ dependencies {
         exclude module: "commons-compress"
     }
     dependenciesCommonJar group: 'com.google.cloud.bigdataoss', name: 'gcs-connector', version: 'hadoop2-2.2.16'
+    dependenciesCommonJar group: 'com.qcloud.cos', name: 'flink-cos-fs-hadoop', version: '1.10.0-0.1.10'
     dependenciesCommonJar 'org.apache.flink:flink-metrics-dropwizard:' + flinkVersion
     dependenciesCommonJar 'org.apache.flink:flink-json:' + flinkVersion
     dependenciesCommonJar 'com.jayway.jsonpath:json-path:2.4.0'