Change how we compute cell capacity and energy

[WardLT]

Overhauls how we compute integral quantities, such as how we compute the capacity and energy change during a cycle.

Our previous implementation relied on integrals split over steps, which relied on correct determination of when a battery is being charged/held/etc and introduced opportunities for numerical issues.

This new version has a simpler implementation (just integrate over the entire cycle) and introduces simple test cases to ensure the batteries are correct.

• [x] Flip the charge conventions everywhere
• [x] Change names such that all related quantities (e.g., capacities) start with the same prefix (ensures they stay together when sorting columns)
• [x] Change names to reflect that positive values represent the amount of energy/charge discharged from a battery. Paul and @victorventuri liked "amp_hours" over "capacity"
• [x] Implement re-using the cumulative charge/energy transferred when computing capacities if available in the raw_data
• [x] Verify that our numbers are close to those in the XCEL datasets
• [x] Make a test case that shows computing capacity on a weird cycle

Fixes #56

[coveralls]

Pull Request Test Coverage Report for Build 8738801161

Details

• 33 of 33 (100.0%) changed or added relevant lines in 4 files are covered.
• No unchanged relevant lines lost coverage.
• Overall coverage increased (+0.4%) to 81.227%

---

Totals
Change from base Build 8622550679:	0.4%
Covered Lines:	649
Relevant Lines:	799

---

💛 - Coveralls