Recently, I worked with BioSSPy which is a signal processing toolbox designed to deal with biosignals demands. It was developed by PIA group (http://pia.lx.it.pt/?page_id=423).
I'm not sure if it is already available online as open source software, but as far as I know PIA group wants to provide it soon.
I suggest you to send a email and explain them all your needs.
you can use labchart pro with ECG module. It finds times and voltages automaticly. I don't think that none of software can be use trustfully for diagnose any disorders from ECG.
But "any disorders that are present in the signal pattern" ... I don't think you're in luck. A lot of electrocardiography people would be reticent to recommend anything which was attempting to classify / categorise all bad beats - at least, anything you can find for free in a biosignal toolbox somewhere.
There are plenty of solutions for individual featural identification (e.g. detect AF vs. no AF) and there is obviously proprietary medical software, but I don't think what you're looking for really exists.
There are several commercially available software packages performing analysis of ECG's from animal models. As my company is one of those providers, I will keep my comment general. ECG's from anesthetized or restrained animal models are fairly easy to analyze, including ECG abnormalities such as arrhythmias. It is much more challenging with ECG's obtained from conscious and free moving animal models, just as it would be challenging for a trained technician. So for the latter, the user should always review and confirm what is reported the automatic detection generated by the software. Of course the amount of data is a factor to consider, if you have days of ECG's, then accepting some level of accuracy might be acceptable.