Add a column name for all first columns in zone data and results

[johpiip]

None of the input zone data files have a name for the first column. Behaviour is the same for result data. Having a name for the first column would make reading and referencing the data easier in R. What the name really is is insignificant. It can be just "group" or "class", or "x" instead of real names like "zone", "municipality", or "mode".

Examples of current behaviour:

• Example of missing first column name in zone data here:

  helmet-model-system/Scripts/tests/test_data/Base_input_data/2016_zonedata/2016.wrk

  Lines 9 to 11 in 12e6c98

  	total sh_serv sh_shop sh_logi sh_indu
  	5 7 0.2 0.1 0 0.1
  	6 5 0.2 0.2 0.1 0

• Example of missing first column in transit_kms.txt:

     dist	time
  metro	14627.57647	20409.43607
  tram	14539.94966	63296.97222
  bus	387704.60918	744000.78360
  ...
  

Improved behaviour:

• 2016.wrk:

  zone	total	sh_serv	sh_shop	sh_logi	sh_indu
  5	7	0.2	0.1	0	0.1
  6	5	0.2	0.2	0.1	0
  ...
  

• transit_kms.txt:

  mode	dist	time
  metro	14627.57647	20409.43607
  tram	14539.94966	63296.97222
  bus	387704.60918	744000.78360
  ...
  

[zptro]

The desired behavior seems to be achieved with index_col=0 parameter in pandas.read_csv() and index_label parameter in DataFrame.to_csv(). The format could easily be changed for output files, but we should then always feed a describing name for the index. Input files are a bit trickier, because it seems we will then drop support for the current format.

[zptro]

@johpiip Should this be considered a Helmet 5 feature? Would it be time to create a development branch?

[johpiip]

@zptro This being Helmet 5 feature sounds like a good idea. Is there a sudden need for this PR or a Helmet 5 dev branch?

[zptro]

@johpiip I just realized that if the user has for some reason added an index label as column name in an input file, the resulting error can be really frustrating to find, at least if the intended zone numbers are very near the integer series 1, 2, 3...