

Thus I estimate that Dash has 218 commits that aren't in Bitcoin. The number may be off slightly due to the repos likely being a bit out of sync with each other.
Not bad, but well under even the paltry 1200 or so attributed to another project that is alleged to be a clone with "no development" so maybe not that impressive either. I'm not sure, what do you think?
I am sure numbers could be checked but I think you are just comparing apples and oranges. In Monero there has been a lot of clean up of the Cryptonote code base done where a few lines of code are changed and then a commit is submitted and certainly other more substantial commits but the small ones sure add up. Dash works differently as new features are added from scratch the amount of code in a commit can be bigger.
It's not apples and organes necessarily. Monero has had some massive commits and new features where work has been done from scratch then added in one commit, including deterministic wallets, the embedded database, encrypted payment IDs, replacement of the RPC subsystem with 0MQ IPC, etc. Those are in addition to improvements and clean up to preexisting code (including preexisting code that was itself previously added by Monero)
I don't really know the best way to measure these things -- I guess different measurements are useful for different purposes -- but when someone posts a comparison of number of commits in Dash compared to another project, then it is important to consider how many of the claimed number of commits actually represent the work of the Dash project and not just the commits that are common to 1000 Bitcoin forks. That number alone may not be the best indication of progress, but at least it is an indication of work being done by Dash.
If you wanted to compare two different projects like this, you should do it in terms of lines of code added
Indeed, I suggested
also looking that metric. And I wasn't even necessarily comparing myself (though I was replying to a comparison) as much as asking about metrics of the actual Dash project work. These sorts of metrics have been published for Monero (for example in our 2014 year-in-review, though that is now quite out of date of course), but I haven't seen them for Dash. Historical metrics would be interesting as well, for example the number of lines of code associated with each of the identified features you mentioned above.