Remove duplicates in the mine_id_msha column

[bendnorman]

Based on the comment in src/pudl/transform/eia923.py the mine_id_msha column should have unique ids:

pudl/src/pudl/transform/eia923.py

Lines 1043 to 1059 in 618c20c

	# If we actually *have* an MSHA ID for a mine, then we have a totally
	# unique identifier for that mine, and we can safely drop duplicates and
	# keep just one copy of that mine, no matter how different all the other
	# fields associated with the mine info are... Here we split out all the
	# coalmine records that have an MSHA ID, remove them from the CMI
	# data frame, drop duplicates, and then bring the unique mine records
	# back into the overall CMI dataframe...
	cmi_with_msha = cmi_df[cmi_df["mine_id_msha"] > 0]
	cmi_with_msha = cmi_with_msha.drop_duplicates(
	subset=[
	"mine_id_msha",
	]
	)
	cmi_df.drop(cmi_df[cmi_df["mine_id_msha"] > 0].index)
	cmi_df = pd.concat([cmi_df, cmi_with_msha])
	cmi_df = cmi_df.drop_duplicates(subset=coalmine_cols)

but they aren't because the cmi_df isn't overwritten.

It's been a while since I've played with this data so I'm not sure what the best option is.