Add normalised code column

[SimonH]

This PR builds on #786, which should be merged first. (or ignored if this PR is approved as a single item)
Using the new same_as values to determine the correct value to add for a new normalised_code column.

Currently the column is only set for non-GB locations. The GB locations will be updated separately at a later date.

I have added a unit test to ensure that all rows have the correct value set in the new column.

[julien-duponchelle]

In your PR the normalized code look like: urn:trainline:public:nloc:eu18

I think the normalized code should not include eu in their name to be futur proof if one day we expand outside the European continent.

[SimonH]

Starting the current normalised identifiers in this file with euXX does not mean that all normalised location identifiers will have the same format. (For example: normalised location identifiers from atoc, which will be going against the GB stations in this file, will be of the form urn:trainline:public:nloc:atocXXX).
When locations from the rest of the world are added, regardless of whether they are added into stations.csv or come from a different source, an appropriate prefix can be assigned at the time to differentiate any numerical clashes.