Fix fuel cost calculation for non-hourly load data

[Bill-Becker]

fuel_cost_per_kwh was erroneously being divided by time_steps_per_hour which is 2 for 30-minute interval and 4 for 15-minute interval loads. The fuel used/consumed calculation already accounts for time_steps_per_hour to get the correct energy used in each timestep, so we need the actual (input) fuel_cost_per_kwh to be preserved to multiply that with fuel used.

This PR avoids dividing fuel cost value by time_steps_per_hour.

[bpulluta]

@Bill-Becker, you want me to add the tests we talked about to check that the hourly and subhourly results are the same? Or is that not neccessary?

[Bill-Becker]

@Bill-Becker, you want me to add the tests we talked about to check that the hourly and subhourly results are the same? Or is that not neccessary?

I think that's a good idea in case we don't have that test coverage already. Just do a simple (short solve time) scenario with Generator that runs and check that that the annual fuel cost is the same for both 15 or 30 minute and hourly.

[bpulluta]

@Bill-Becker, you want me to add the tests we talked about to check that the hourly and subhourly results are the same? Or is that not neccessary?

I think that's a good idea in case we don't have that test coverage already. Just do a simple (short solve time) scenario with Generator that runs and check that that the annual fuel cost is the same for both 15 or 30 minute and hourly.

@Bill-Becker Okay, added to the tests. Let me know if you'd like me to put it into a separate test section called Subhourly or something. Seems like you fixed the discrepancy with your solution

[bpulluta]

@Bill-Becker did this get integrated separately? Wondering if this was fixed elsewhere

[Bill-Becker]

@Bill-Becker did this get integrated separately? Wondering if this was fixed elsewhere

Hmmm. I vaguely remember you opened up a second PR and that one took precedence over the first one, but I don't know how it would have gotten fixed if neither were merged?!

[bpulluta]

@Bill-Becker this pr replaced the previous one that was deleted. We might need to open this and review. I approved this and is ready for merging after your review.

[Bill-Becker]

@bpulluta Can you take a stab at making the Offgrid test that you added a little "easier". It's passing for me locally, but failing on GitHub Actions. It takes a long time to solve, and I'm wondering if that's causing some inconsistency in the results on GitHub Actions with the limited computational resources it has to solve.