If the Arduino is anything like a PIC µC then you have no hope of sampling at 44KHz. Most simple µC have quite a slow sampling rate (like 100's of samples per second).
If you want faster then you'd be looking at using something like a dsPIC which has an actual audio grade ADC in it, or use an audio ADC externally that can send I²S data to a µC that is fast enough to respond to it.
I have done some similar work recently while designing a digitally controlled amp.
I had the output of the first stage of the amp going into an analog input on the controlling PIC to then control a bargraph of LEDs for a simple VU meter.
For an output from a PC soundcard you're probably looking at around 1 to 2 volts voltage swing. For my system I wasn't too fussed about frequency and such - just pure peak amplitude - so I passed the signal through a small shottky diode first to trim off the negative voltages. This simplified my design a whole lot.
I am also designing a small frequency analyzer at the moment, and am looking at having selectable op-amp based band-pass filters based around this design: http://www.wa4dsy.net/robot/bandpass-filter-calc which so far has given quite good results. I am varying some of the resistor values by a combination of digital pots and analog multiplexers.
I would certainly recommend at least protecting your analog input(s) with op-amps to limit the maximum voltage they get - just in case. You don't want a voltage spike blowing up your Arduino now do you? Easier to replace a blown op-amp.
And as for a signal for testing? There are many free signal generators for the PC available for download if you do a little google for them. They will let you select waveform, frequency, amplitude, phase, etc. Even allow summing of waveforms to give new waveforms if you're lucky.
You can even use your PC soundcard as a rudimentary scope as well with the right software and a small home-made probe. There is software and designs around for this too on the net.
Oh, and remember to isolate different stages / voltage levels with capacitors in the audio signal. As a rule of thumb, if I am changing PSU voltage levels, I always introduce a capacitor to isolate the stages. So, I had one on the input signal, one on the stage 1 -> stage 2 (+/-5V to +/-12V power supply), one on the stage 1 -> analog input, and one again on the output. It pays to take no chances with stray DC offsets wandering into the wrong part of the circuit.
Best Answer
This probably fits dsp.stackexchange.com better.
This is the most simple FT that you should understand. The basic rule is: a sine of frequency f0 in time domain is a delta at f0 and -f0 in frequency. You can do this theoretically, as shown in http://mathworld.wolfram.com/FourierTransformCosine.html.
Those peaks in your graph are consistent with this. However, it isn't true that there is a peak in at 90Hz. The peaks should be at 10 Hz and -10 Hz. There are two possible approaches here:
a) Use fftshift to correct this issue. This will take each frequency to its real place.
b) Assume the signal is real, so the spectrum is symmetric. It is true in this case, but it doesn't have to. You could just plot the first half of your fft, i.e. from 0 Hz to 50 Hz.