The issue with this sort of analysis is that it is incredibly expensive. Knowing if each test "could have" killed a mutant requires pitest to run all tests covering a mutant against each mutant, which makes it impractically slow for day to day development. For this reason it is not something that will ever be implemented out of the box in pitest.
Less expensive information that could be captured is a list of all killing tests. A metric of "104 out of the 1000003 available tests were required to kill these mutants" could be produced from this.
With a bit of work, your original lax/slack tests metric can be calculated though. Pitest has a "not really supported or documented" `fullMutationMatrix`mode which was added for academic use. When activated, pitest will keep running tests after a killing one is found. Details of the other tests run are available in the XML output (but not in other formats). It should be possible to extract the data you need from this.
Henry