Refactor ZoningSystem initialisation method

[wsp-mbuckley]

Remove zone descriptions, internal zones and external zones parameters from ZoningSystem init and instead include this data as columns in the dataframe. The dataframe should contain the following columns:

• zone_id: mandatory zone ID integers
• zone_name: optional zone names
• zone_descriptions: optional zone descriptions

Then can contain any number of additional boolean columns for flagging different areas or types of zones, descriptions of these columns can be included in the zoning system metadata.

zone_id	zone_name	internal_tfn	external_tfn	north_england	south_england
1	Manchester	1	0	1	0
2	London	0	1	0	1

Add validation of the dataframe which does the following:

• Copy dataframe to avoid editing it inplace
• Drop any possible multilevel column names or indices which aren't needed
• Normalise column names, convert to lowercase, strip end whitespace, replace spaces with underscores
• Check for mandatory columns, just 'zone_id'
• Check for 'zone_name' and 'zone_descriptions' columns
• Check for any flag columns and make sure they're booleans e.g. 'tfn_internal', 'tfn_external'

Add 2 property methods to return the zone names and descriptions lookups (probably a Series for each with the zone ID as the index).

Add method for getting any subset of zones based on the boolean flag columns #5

[wsp-mbuckley]

Including all the zone information in the dataframe would make the save and load methods simpler.

[wsp-mbuckley]

Resolved in #18