Compute capital gains using FIFO model for FHorz agents

[MichaelTiemann]

These changes use a simple last-in/first-out model of stock acquisition and disposal. If an agent should choose to sell stock before max acquisition period is upon them, we presume it's hardship due to a shock, and that they basically liquidate stock acquired the previous period.
For stocks sold peak acquisition, we calculate the distance from the agent's age to the peak and presume they are selling stock purchased proportionally prior to the peak acquisition period. This could be improved by noticing whether there's a wide peak (in which case we should center the peak). But the general concept is to capture the taxes that fall on a symmetric buy/sell pattern.
We have to adjust the initial price of the stock because older agents who acquired the stock when they were younger buy in at a very low cost (relatively speaking). The price is selected so that the stock value balances to 1, meaning younger agents pay above 1 and older agents paid below one when they "bought" the stock they hold.

[MichaelTiemann]

I'm going to add the open question as to whether the stock should be presumed a "growth" stock (meaning price appreciates >= risk-free return rate) or "dividend" stock (as it is presumed to pay a rich dividend). A likely middle-ground is to treat the dividend as the stocks up-side, and to imagine that its share price grows at the rate of risk-free return.

Other thoughts welcome...