Revert "feat!: export crypto primitives (#1728)"

[fryorcraken]

This reverts commit 7eb3375

The original PR states:

In some cases users don't want to encrypt whole payload but only some part of it to have additional features.

I would argue the opposite. If a user has a key to decrypt or encrypt, then Waku Message v1 should be used to encrypt the whole message.

By only encrypt part of the messages, we open the door for developers to encrypt metadata and leak them. Which against why we are building Waku in the first place.

Also, within days of exposing these primitives, I got a developer confused on whether they should use them.

[github-actions]

size-limit report 📦

Path	Size	Loading time (3g)	Running time (snapdragon)	Total time
Waku core	78.48 KB (0%)	1.6 s (0%)	217 ms (+85.58% 🔺)	1.8 s
Waku Simple Light Node	241.3 KB (0%)	4.9 s (0%)	322 ms (+52.75% 🔺)	5.2 s
ECIES encryption	73.1 KB (+0.19% 🔺)	1.5 s (+0.19% 🔺)	173 ms (+2.17% 🔺)	1.7 s
Symmetric encryption	73.11 KB (+0.22% 🔺)	1.5 s (+0.22% 🔺)	190 ms (+49.47% 🔺)	1.7 s
DNS discovery	120.91 KB (0%)	2.5 s (0%)	314 ms (+144.68% 🔺)	2.8 s
Privacy preserving protocols	126.01 KB (0%)	2.6 s (0%)	138 ms (-16.41% 🔽)	2.7 s
Light protocols	76.06 KB (0%)	1.6 s (0%)	193 ms (+45.64% 🔺)	1.8 s
History retrieval protocols	74.97 KB (0%)	1.5 s (0%)	153 ms (-6.04% 🔽)	1.7 s
Deterministic Message Hashing	5.65 KB (0%)	113 ms (0%)	106 ms (+566.3% 🔺)	219 ms

[adklempner]

Does this change affect the notes example PR? waku-org/examples.waku.org#286

[fryorcraken]

Does this change affect the notes example PR? waku-org/js-waku-examples#286

I am stating that we do not need these API for the notes example: waku-org/examples.waku.org#291

I do need to triple check.

[weboko]

This and we should keep exposing those methods to prevent unnecessary packages being added by consumers in case they would try to do encryption, signing.

[weboko]

Closing this for now, let's change it in future based on feedback from people.

[weboko]

#1955