A Simple way to validate Data sources connectivity in Spark Job

When more and more data sources are involved, current E2E test is not scalable for pure connectivity test.
A separate spark job to simply try to **load all registered data sources into data frame** can help to

- **Enhance Test and Engineer Experience**
- Scalable data format supportive check
- Validate data connectivity for customized data source
- Further Usage in Feature Store UI / Data Platform*
- Data Source Healthy Telemetry (Daily job)
- Data Visualization Sampling/Distribution

To achieve this, we may need to have:

- **A list of sample data source** which covers every support data formats (can be a config file synced to sample feature registry) with
- data path
- credential pointer (to a centralized credential storage, e.g. Key Vault)
- rules* (customized rules to make sure data source meets requirements)
- **Credential Storage**
- key or token
- access type*: e.g. admin / read / write...
- access level*: e.g. single file / folder; table / storage...
- **DataSourceCheckJob(ss, Seq[dataSourceDef])**

"*" : means nice to have & low priority