The differences between American Standard English and AAVE are primarily grammatical, not lexicographical. So an AmE-AAVE dictionary wouldn't be very thick.

And Leetspeak, txt-speak, and AAVE are not equivalent entities. Leetspeak is an alternative alphabet, txt-speak is a system of abbreviations and slang, and AAVE is at minimum a dialect.

Leetspeak can be classified as a dialect too, if you'd like, it has its own words (leet, pwnd, haxor, etc.) and you can find grammar differences if you look hard enough[1].

[1]https://en.wikipedia.org/wiki/Leet#Grammar

That's pretty unconvincing. Despite the heading "Grammar", it mentions differences that are lexicographic or merely orthographic in nature. The only real grammatical process mentioned is "changing its grammatical usage to be deliberately incorrect" which has some obvious problems. "All your ___ are belong to ___" is an idiomatic expression in Internet Standard English, but that does not a grammar rule make.

You haven't articulated any real objective principle to distinguish when two ways of talking are the same language. The decision seems to be based on politics (in this case politics of race)

Naturally, there is no such objective principle for me to articulate. Everyone speaks according to the language center of their own brain. Some ways of speech are more different than others; some are so different that communication is hard without speaking slowly, and some still are so different that communication is impossible and you must resort to pointing (and even pointing is not universally meaningful). At some point along the line we call it a different language, but it's not representative of any sharp distinction that exists in reality.

Asking a linguist "are these the same language" is as useful as asking a biologist "are these the same species"; it all depends on what you want to use them for.

(Although I don't particularly see the relevance of that to whether or not leet has a substantially distinct grammar from English, of which I remain unconvinced.)

>So should there be AAVE-English academic dictionaries with entries like "ain't: are not", etc.?

This is already the case in Standard English dictionaries: http://www.merriam-webster.com/dictionary/aint

>What about l33tsp34k and txtng spk, u wnt 2gt dctnrz 4 it 2?

Actually, given how often I see it, I wouldn't be surprised if "U" was the standard English spelling for the second-person pronoun (or at least an acceptable variant), by the end of this century. It wouldn't be the first time an English pronoun's spelling collapsed down to a single letter ("I" was originally "Ic").

U is logically parallel to I, and is accepted in (India) Indian English.