These are subtly needed but not obvious from the previous result processors because they always processed regardless of results. Need to ensure sanity checks in tests on how we handle a key that isn't there as well.

Matching this to #2095 and I'll move others here separately.

@NickCraver, working on `BITFIELD`/`BITFIELD_RO` and I realized that several of the new commands we've been adding are write-commands and therefore won't work on replicas, opening this to fix the command-map in Message.cs (pretty sure this is the right place but LMK if I'm offbase here) will need to push an update to #2095 #2094 to reflect this as well.

Co-authored-by: Nick Craver <nrcraver@gmail.com>