Based on what can be gleaned from the website you linked you ought not to exceed the phase current specified. This is how I see but if you can find proper data sheets, they might tell a different story.
Firstly look at the torque-speed curve (linked on the page): -
If you do the math, you'd calculate that the maximum mechanical output power is about 1 watt (100 rpm and 0.1 N.m = 1.047 watts). This is about the same at 200 rpm ( 200rpm and 0.06 N.m = 1.26 watts). At 400 rpm output power is 0.837 watts hence, you can see the max output power is 1.26 watts.
With a coil current of 0.31 amps per phase and a coil resistance of 38.5 ohms (as stated), the power (heat generated not mechanical power) is 0.31^2 x 38.5 watts = 3.7 watts and this means your stepper motor can be running quite hot.
Having said that, this "apparent" inefficiency (some simplification and assumptions made here) of about 25% will not be at optimum mechanical output. At optimum power output ( I reckon about 100rpm), the power in will be about 2.1 watts assuming a peak efficiency of about 60%. This is about "normal" for steppers of this type.
So, if you are always going to be running about 100 rpm, the current into the coils will be lower than 0.31 amps BUT, the details in the link are really unclear about this so caution should be taken.
Conclusion - I don't think you can dare run the coils at more that 0.31 amps based on what the specification says. I recommend finding out more about the device. Try looking at one from a regular supplier i.e. one that has a proper pdf data sheet and deciding what information that data sheet provides that this one doesn't.
You will be getting the HOME pulse every 4 steps in full-step, or every 8, 16, or 32 steps in microstepping modes. In microstepping, the state table contains "values" for the current to be sent in each of the two windings to set the rotor in a particular position relative to the full step (which doesn't need a state table). IIRC this is the position with the highest holding torue.
The microstepping waveform vaguely resembles one period of a sinusoid; in the table you've posted last column shows the "position" or angle corresponding to the microstep on this sinusoid.
Best Answer
One way to increase the resolution so that you can make steps close to an exact one degree is to use a microstepping driver.
That kind of driver works by PWM-modulating the current to the motor coils in a ratio that allows the rotor to stabilize in a position between the poles. You can't really increase the resolution by a huge factor, despite the driver stepping in (say) 128 or 256 microsteps but it might be good enough for you.
If you had 8 microsteps on a 1.8 angular degree motor, then each step would represent 0.225 degrees, and every 40 microsteps (9°) you'd be back to an exact full step.
You can make your own, or there are commercial suppliers such as Gecko.