Add some AD suggested wording to content coding decoding discussion

[LPardue]

I found this a little difficult to integrate in and hence its quite a mouthful of some sentences. However, I think it captures what was described in the issue.

Fixes #3356

[mikewest]

I'm fine with your text, but I took a stab at changing some of the words. Hopefully it transforms your text in a way that improves scannability "without loss of information"? :)

If you prefer your original text, land it!

[mikewest]

WDYT about this suggestion?

Suggested change

	information", that doesn't necessarily mean a byte-for-byte equivalence. Many
	registered content codings do provide equivalence but there is no requirement
	for it; it remains a possibility that decoding could produce a different byte
	sequence. In order to avoid unintended validation failures, care is advised when
	selecting content coding for use with `Unencoded-Digest`.
	information", that doesn't necessarily mean a byte-for-byte equivalence. It's
	entirely possible for content codings to perform semantically-meaningless
	transformations that nevertheless result in a decoded byte sequence that does
	not exactly match the original unencoded representation.In order to avoid
	unintended validation failures, care is advised when selecting content coding
	for use with `Unencoded-Digest`.

[LPardue]

I worry that this maybe makes the risk sound too large. I think Julian identified one content coding that behaves that way, and I don't think its widely used that. I hope that implementers of the spec can just assume things like gzip, brotli etc can be used without issues. I'll make a hybrid change that incorporates some of your proposed text.

[LPardue]

PTAL a look at the latest version after I pushed a commit.

[MikeBishop]

Either version works for me.

[MikeBishop]

Would this help to further scope the size of the risk? It's worth calling out for completeness, but in practice most will be byte-for-byte unless they're media-type-specific.

[LPardue]

With the two applied suggestions, I think that it strikes a decent balance on identifying the scope of risk, without being overlying rabbitholing

[LPardue]

In the interest of moving forward, I'll merge this and cut a new release. We can continue to wordsmith up until publication if its really necessary.

Thanks for the input!