Update SCIP data

[corviday]

This is mostly a big data update; code changes are minimal.

Data is now available for a much larger area, including Haida Gwaii, Vancouver Island, and most of the mainland BC Coast.

Please poke at the demo a little and see if you can find anything broken. Haida Gwaii is the only entirely new top-level region, so it's a good place to poke around.

Resolves #81

[QSparks]

LGTM! Not much to review, just spotted a couple of typos.

Out of scope: I tested the demo and noticed that if you first select a watershed that includes a downstream path, then change to one that does not, the previously highlighted path is not removed from the map.

[corviday]

Test notes from @mschnorb :

• The option to select a logarithmic scale for the color legend is not working; i.e. it doesn’t work for streamflow related metrics when it would be most useful, even though the minimum value in the data is >0.
• Need to provide a minimum and maximum for the lowQ05_jjason indicator on the color bar
• I did also note another interesting artefact on both the demo and operational version: it seems that in the map the hazard grids plot in two separate locations on the globe - in BC and also over the arctic (see attached figure). I guess I have never zoomed the map out that far before.

[corviday]

Response to bugs @mschnorb found:

1. Logarithmic scaling was disabled due to 0 values in the flow indicators, which would cause an error if we attempted to construct a logarithmic scaled image of them. These errors do not seem to be in the new data (hooray!) so I have re-enabled logscaling.

2. I had forgot to configure lowQ05_jjason. This has been remedied.

3. I think this is a side effect of using the BC Albers map; leaflet gets weird if you zoom outside of the place where the map coordinates are defined. This is not a priority to fix, and has not been fixed.