Masking Strategies: Handle Null Values and Non-Strings

[pattisdr]

Closes: #2391

Code Changes

• Hash, Hmac, and AES Encrypt masking strategies adjusted to better handle null values and non-strings (here we try to coerce to a string instead of failing)

Steps to Confirm

• list any manual steps for reviewers to confirm the changes

Pre-Merge Checklist

• All CI Pipelines Succeeded
• Documentation:
  • documentation complete, PR opened in fidesdocs
  • documentation issue created in fidesdocs
• Issue Requirements are Met
• Relevant Follow-Up Issues Created
• Update CHANGELOG.md

Description Of Changes

Several of our masking strategies are set up to expect values to be strings. There are some early exits if the input value is None, but since most inputs are now a list of values, even a list of one value, so that's not likely to be True.

If an array of values are passed in, if the original index is None, do nothing to that index. Also try to coerce the original value to a string.

[pattisdr]

@adamsachs wanted to get your thoughts on the strategy here

[adamsachs]

this looks good to me @pattisdr! nice quick fix.

the None handling is exactly what i'd expect, and that's the most important piece.

the non-string-type coercion is a tricky one, and i agree with the approach of trying -- essentially, why not? my only question is whether there's anything we can do to make a somewhat-expected type coercion error from the DB clearly communicated to the end user. maybe it's already going to be clear in how it's set up now, i'm not sure -- just wanted to raise point up to ensure it's considered.