Implement operator class parameters

PostgreSQL provides set of template index access methods, where opclasses have
much freedom in the semantics of indexing.  These index AMs are GiST, GIN,
SP-GiST and BRIN.  There opclasses define representation of keys, operations on
them and supported search strategies.  So, it's natural that opclasses may be
faced some tradeoffs, which require user-side decision.  This commit implements
opclass parameters allowing users to set some values, which tell opclass how to
index the particular dataset.

This commit doesn't introduce new storage in system catalog.  Instead it uses
pg_attribute.attoptions, which is used for table column storage options but
unused for index attributes.

In order to evade changing signature of each opclass support function, we
implement unified way to pass options to opclass support functions.  Options
are set to fn_expr as the constant bytea expression.  It's possible due to the
fact that opclass support functions are executed outside of expressions, so
fn_expr is unused for them.

This commit comes with some examples of opclass options usage.  We parametrize
signature length in GiST.  That applies to multiple opclasses: tsvector_ops,
gist__intbig_ops, gist_ltree_ops, gist__ltree_ops, gist_trgm_ops and
gist_hstore_ops.  Also we parametrize maximum number of integer ranges for
gist__int_ops.  However, the main future usage of this feature is expected
to be json, where users would be able to specify which way to index particular
json parts.

Catversion is bumped.

Discussion: https://postgr.es/m/d22c3a18-31c7-1879-fc11-4c1ce2f5e5af%40postgrespro.ru
Author: Nikita Glukhov, revised by me
Reviwed-by: Nikolay Shaplov, Robert Haas, Tom Lane, Tomas Vondra, Alvaro Herrera

doc/src/sgml/intarray.sgml

@@ -265,7 +265,7 @@
   </para>
 
   <para>
-   Two GiST index operator classes are provided:
+   Two parametrized GiST index operator classes are provided:
    <literal>gist__int_ops</literal> (used by default) is suitable for
    small- to medium-size data sets, while
    <literal>gist__intbig_ops</literal> uses a larger signature and is more
@@ -274,6 +274,25 @@
    The implementation uses an RD-tree data structure with
    built-in lossy compression.
   </para>
+   
+  <para>
+   <literal>gist__int_ops</literal> approximates integer set as an array of
+   integer ranges.  Optional integer parameter <literal>numranges</literal> of
+   <literal>gist__int_ops</literal> determines maximum number of ranges in
+   one index key.  Default value of <literal>numranges</literal> is 100.
+   Valid values are between 1 and 253.  Using larger arrays as GiST index
+   keys leads to more precise search (scan less fraction of index, scan less
+   heap pages), but larger index.
+  </para>
+   
+  <para>
+   <literal>gist__intbig_ops</literal> approximates integer set as a bitmap
+   signature.  Optional integer parameter <literal>siglen</literal> of
+   <literal>gist__intbig_ops</literal> determines signature length in bytes.
+   Default signature length is 16 bytes.  Valid values of signature length
+   are between 1 and 2024 bytes.  Longer signatures leads to more precise
+   search (scan less fraction of index, scan less heap pages), but larger index.
+  </para>
 
   <para>
    There is also a non-default GIN operator class
@@ -293,8 +312,8 @@
 -- a message can be in one or more <quote>sections</quote>
 CREATE TABLE message (mid INT PRIMARY KEY, sections INT[], ...);
 
--- create specialized index
-CREATE INDEX message_rdtree_idx ON message USING GIST (sections gist__int_ops);
+-- create specialized index with sigature length of 32 bytes
+CREATE INDEX message_rdtree_idx ON message USING GIST (sections gist__int_ops(siglen=32));
 
 -- select messages in section 1 OR 2 - OVERLAP operator
 SELECT message.mid FROM message WHERE message.sections &amp;&amp; '{1,2}';