Implement strategy pattern for file reading

[htmlboss]

At the moment, the tool only supports indexing of netcdf files. Some datasets use the GRIB file format. This PR implements some architecture to enable support for multiple file types to be indexed.

• Implemented CRTP to enable easy addition of logic to read different file types other than netcdf.
• Refactor out netcdf-specific reading logic from DatasetDesc class into NCFileReader class.
• Create empty Grib2FileReader for future implementation.
• Reorganized source files into more descriptive directories.
• Minor cleaning of erroneous keyword usage.

[x] Unit tests pass [x] Tested

[htmlboss]

Before merging: add conda environment file.

[htmlboss]

Here is an overview of what got changed by this pull request:

Issues
======
- Added 1

See the complete overview on Codacy

[htmlboss]

Conda environment file added.

I've tested the netcdf-side of the tool and the generated table is correct.