Introduce unique identifiers for lines and derubricate line numbers to "folk names"

[albertocottica]

With Miccolis, line numbers are only loose indicators of what the buses will actually do. I propose to

1. add a field that uniquely identifies a "line" as a vector of stops. Two runs belong to the same line only if they are described by the same vector. According to this rule, line 14 is in reality several different lines, as the routes vary, according – for example – to whether we are looking at a weekday or weekend.
2. include the present line field not as an identifier, but as a "folk descriptor" – it is what the passengers see on the bus! By doing this, we can complete the files containing the timetables. The price we pay is that we need to bring each of these lines (in the new way) onto OSM. This would not be so bad if we knew (1) the location of all poles (Palo) and (2) the indication of the poles in Miccolis's timetables were reliable. These two items translate into questions we can ask them!

[piersoft]

Alberto see http://87.230.17.190:7777/map the line number is already visible. we need line name (route_long_name)

[albertocottica]

I am afraid it is not only that. We will need to create more records, one per each _route_longname.