Rename hnsw.init_k to lantern_hnsw.init_k

[var77]

Rename hnsw.init_k parameter to lantern_hnsw.init_k to match the naming convention we have for lantern_hnsw.ef_search parameter

[github-actions]

Benchmarks

metric	old	new	pct change
recall (after create)	0.740	0.740	-
recall (after insert)	0.762	0.746	-2.10%
select bulk tps	488.520	472.036	-3.37%
select bulk latency (ms)	15.345	16.009	+4.33%
select bulk latency (stddev ms)	3.092	2.973	-3.85%
create latency (ms)	1209.939	1216.015	+0.50%
insert bulk tps	11.154	11.020	-1.20%
insert bulk latency (ms)	89.651	90.740	+1.21%
insert bulk latency (stddev ms)	3.534	2.824	-20.09%
disk usage (bytes)	6348800.000	6348800.000	-

[codecov]

Codecov Report

Merging #219 (2b926ea) into main (8110b58) will not change coverage.
The diff coverage is 100.00%.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #219   +/-   ##
=======================================
  Coverage   77.92%   77.92%           
=======================================
  Files          20       20           
  Lines        1468     1468           
  Branches      370      370           
=======================================
  Hits         1144     1144           
  Misses        160      160           
  Partials      164      164           

Files	Coverage Δ
src/hnsw/options.c	87.93% <100.00%> (ø)

[Ngalstyan4]

should this only be done when pgvector is also installed?
we only use lantern_hnsw if hnsw is already taken by pgvector

[var77]

I did this change to match the naming with the other variable we have lantern_hnsw.ef_search . As that also has the prefix lantern_, we can decide to use hnsw or lantern_hnsw (or use lantern_hnsw only when pgvector is installed) what do you think?

[Ngalstyan4]

hm, lantern_hnsw was supposed to only be for pgvector compatibility.
hnsw must exist otherwise. (lantern_hnsw can exist otherwise as well, but hnsw MUST exist)
So, this should be changed accordingly