Post-1054 follow-ups

[ariard]

We have likely few redundant checks in send_htlc (see #1054 (comment)), cleaning them would ease the refactoring of our fee(s) accounting helpers (next_local_commit_tx_msat, get_inbound_htlcs, build_commitment, etc) as consumed by the affordance checks at HTLC sending and reception. There is at least fixing the check in update_add_htlc by using the correct anticipated value instead of value_to_self_msat (see #1054 (comment)). And likely adding test coverage, especially w.rt handling of dust/dust_limit_satoshis.

[ariard]

#1054 (comment)

[valentinewallace]

Also #1054 (comment)

Disregard, it was resolved I think

[ariard]

There is still test coverage for dust exposure at inbound update_fee to add but note it was added before 1054.