Pesticide application thresholds

[bradduthie]

Include the following features into an R package update:

• An argument (e.g., pesticide_threshold) that sets a pest per cell density threshold for farmers to apply pesticide. I'll allow this argument to be either a scalar applied to all farmers or a vector of the same length as farms (so you could, in theory, give different farmers different thresholds -- probably worth it to write it this way in the long run).
• An argument (e.g., pesticide_threshold_delay) that sets the delay before pesticide application is started (and stopped?)

These will allow users to give farmers the flexibility to not apply pesticide until pest population density is sufficiently high, then stop applying it once pest density decreases.

[bradduthie]

This is implemented, but I'm not entirely satisfied with how pest_dense works in the sim_farming function. If the delay is set to zero, then the delay is set to zero and the count does not go anywhere -- meaning that pesticide continues to be applied even if the pest population drops below the threshold density (see densities.c line 56). If it drops below the threshold, this delay really should be at least 1, so maybe just changing to delay_count[i] = delay[i] +1; is sufficient?