I think you need to take a cold hard look at the numbers.
Say a professor graduates 10 students during her career. Only one is needed to replace her, so we start with a 10% chance. We don't have a good idea of what an acceptable job is - national labs, industrial labs, SLACs, whatever, so lets double that. So maybe its one out of five with large error bars.
About 2000 people graduate with physics PhDs per year, so that means ~400 acceptable jobs. There are about 12000 physics bachelors awarded annually, so the odds start at 400/12000 or just over 3%.
We can already quibble with these numbers, but it sets the scale - once he gets a physics BS or equivalent (and he isn't even there yet) the odds are in the single digits of percents. Most likely the low single digits.
Am I being discouraging? It's not me. It's the numbers. The same problems crop up is one wants to be a professional baseball player, or oboist, or actor. You have a large number of people competing for a small number of slots - and I haven't even discussed the OPs specific situation. This is the
starting point.
The OP now has two hurdles beyond this to overcome:
1. He needs financial support, but most of the available financial support was used up for gis first degree. Nobody disagrees that the OP has a ways to go before he is prepared and a ways to go before he is competitive. The problem is he needs someone else to pay for it.
2. A 3.2 is not competitive.
Now, the OP hopes to get a higher GPA, and do well on the GRE and get great letters etc. But so do the other 12000. And the aspiring athletes and dancers and bassoonists.
Given these numbers, I would say the OP's priorities should be:
- Define success more broadly
- Come up with a plan, not just a list of hopes. That plan needs off-ramps, e.g. "If I don't have a GPA of X by time Y, it's time to do something else."
- Work on a credible financial plan to do this that doesn't rely on other people paying the bills.