Speed up T64 decompression via dynamic dispatch

[Algunenano]

Changelog category (leave one):

• Performance Improvement

Changelog entry (a user-readable short description of the changes that goes into CHANGELOG.md):

Speed up T64 decompression via dynamic dispatch

Details

Comes from #90043.

Easily 2x+ as fast (with local CPU with 512BW):
SELECT count(n) FROM codec_seq_UInt64_T64 WHERE ignore(n) == 0 LIMIT 20000000 SETTINGS max_threads=1;

• Before: 0.0.0.0:49000, queries: 200, QPS: 16.504, RPS: 330077530.024, MiB/s: 2518.292, result RPS: 16.504, result MiB/s: 0.000.
• After: 0.0.0.0:49000, queries: 200, QPS: 37.334, RPS: 746685862.706, MiB/s: 5696.761, result RPS: 37.334, result MiB/s: 0.000.

[clickhouse-gh]

Workflow [PR], commit [c735276]

Summary: ❌

job_name	test_name	status	info	comment
Build (amd_compat)		failure
	Cmake configuration	failure	cidb
BuzzHouse (amd_debug)		failure
	Logical error: 'Inconsistent AST formatting: the query:	FAIL	cidb
BuzzHouse (amd_ubsan)		failure
	/home/ubuntu/actions-runner/_work/ClickHouse/ClickHouse/src/IO/VarInt.h:32:5: runtime error: store to null pointer of type 'char'	FAIL	cidb

[Algunenano]

[rschu1ze]

Nice optimization! @Algunenano