Add new sectors to NAICS crosswalks upon crosswalk load

[bl-young]

Needed for external activitytosectormapping files that may include new NAICS (e.g. 7 or 8 digit NAICS) not currently in the crosswalk.

Crosswalk could be appended locally after loading, perhaps in load_crosswalk()

flowsa/flowsa/common.py

Lines 59 to 76 in aaf5403

	def load_crosswalk(crosswalk_name):
	"""
	Load NAICS crosswalk between the years 2007, 2012, 2017
	:return: df, NAICS crosswalk over the years
	"""
	cw_dict = {'sector_timeseries': 'NAICS_Crosswalk_TimeSeries',
	'sector_length': 'NAICS_2012_Crosswalk',
	'sector_name': 'NAICS_2012_Names',
	'household': 'Household_SectorCodes',
	'government': 'Government_SectorCodes',
	'BEA': 'NAICS_to_BEA_Crosswalk'
	}
	fn = cw_dict.get(crosswalk_name)
	cw = pd.read_csv(f'{datapath}{fn}.csv', dtype="str")
	return cw

Currently that crosswalk will only include activitytosectormapping files and NAICS within the flowsa repo

[matthewlchambers]

Functionality to add such new "NAICS" codes (or any other industry codes that aren't actual NAICS codes) is in the seea-recursive branch. I needed it to account for emissions from waste decomposition in accordance with SEEA, which says these emissions should be attributed to "accumulations," not to the (current activities of) the waste management industry.

flowsa/flowsa/naics.py

Lines 33 to 37 in fc14218

	The top industry_spec dictionary may also include a key 'non_naics', where
	the associated value is a non-NAICS "industry" or list of such "industries"
	that should be included in the set of industries that can be mapped to.
	In this case, the user will need to supply their own crosswalk which maps
	activities to that industry.