Add "click to adjust" for use with color calibration paper

[jywarren]

Planning this out...

1. clicking on a region (maybe later selecting a region by dragging...) and
2. filling out two coefficients for the reflectivity of the a. visible and b. infrared channels as described in this post

A calibration image might look like this (esp. if we had an overlay as i've doodled on it) and might even be of a printed sheet that you line up with the guidelines:

(looking at selection of targets in this post)

This could be repeated as many times as we'd like but we might start with 2 targets minimum as Ned mentioned in the linked post above. Starting with 2 targets is a good initial step because the values could be used to map (or stretch, see this code) rather than doing a linear regression. Just to take things one step at a time.

So for each of 2 points clicked on, we would have values as follows (data from this post):

	Visible reflectivity	Infrared reflectivity
Target paper 1 (printer paper)	0.86696300	0.90032700
Target paper 1 (tar paper)	0.04748605	0.05665055

But by clicking on the image, we've collected a matching set of data through this particular camera, so we can see how different it's reading of these targets is (here, i just made up some data):

	Visible reflectivity	Visible pixels in photo	Infrared reflectivity	Infrared pixels in photo
Target paper 1 (printer paper)	0.86696300	0.8	0.90032700	0.9
Target paper 1 (tar paper)	0.04748605	0.2	0.05665055	0.3

Now, we can use the map() function above to adjust any value in the image, in two steps, one per channel:

1. every Visible pixel gets mapped from a range of 0.2-0.8 (see table above) to a range of 0.04748605-0.86696300
2. every Infrared pixel gets mapped from a range of 0.3-0.9 to a range of 0.05665055-0.90032700

(in a future version we could use a linear regression as Ned did, instead of just a simple map: https://github.com/Tom-Alexander/regression-js)

the visible and infrared values are then displayed and made usable in the usual NDVI equation of (B-R)/(B+R) (or equivalent)

Finally, this whole thing could be in a tray that opens above the image. Later, it could be used in Image Sequencer as a step like this:

https://sequencer.publiclab.org/examples/#steps=dynamic{red:r*0.8|green:g|blue:b*0.4|monochrome%20(fallback):r%20%2B%20g%20%2B%20b},ndvi{filter:red}

[Deepthi562]

@jywarren is the issue open or closed? kindly close it if it's done.