[lmi] Experimental implementation of non-MEC solves

[Greg Chicares]

Use 20090217T0223Z HEAD to test an experimental implementation of
non-MEC solves.

We aren't yet sure what the user interface should be, so this
implementation temporarily hijacks another solve's interface.
Type "idiosyncrasyN" in the "Comments" field to make the solve
for tax basis perform a non-MEC solve instead. Although this
permits a target CSV and a target duration to be specified, those
two fields are ignored for non-MEC solves: the "target" is simply
to avoid a MEC for the lifetime of the policy.

A non-MEC solve probably doesn't make sense for loans. It might
for withdrawals, but we aren't yet sure how to handle that. For
now, solve for either level specified amount or level premium.

We're focusing on CVAT today. Major changes will be needed for
GPT MEC testing to reflect recent legal interpretations.

This non-MEC solve replaces the old feature that increases the
specified amount to avoid a MEC. That old feature will soon be
removed. It wasn't very practical, because requiring a (large)
increase every seven years presumes future insurability. And the
way we implemented it in 2002 left much to be desired: it fails
to increase the initial specified amount when necessary, and it
doesn't work at all for group carveout.