Improve aggregation and status performance (no conditions).

[joaander]

Description

• Use signac 2.2.0 cached_statepoint in aggregators.
• Cache the list of job ids while _buffered and avoid filesystem checks during aggregation and status check loops.

This first pass is tested against workflows with no pre/post conditions. I have plans to optimize [pre/post].isfile in a later PR.

Motivation and Context

Significantly improve the performance of common flow operations, especially workspaces with many jobs and aggregates.

Here are benchmark results performed on the Great Lakes scratch filesystem with 100k jobs.
signac 2.2.0, flow main:

Command	operation	groupsof	groupsof_sort	groupby	groupby_sort
import	0.000204	0.000206	0.000209	0.000212	0.0002
instantiate	0.00537	46.0	177.0	173.0	175.0
status	3.5	45.4	173.0	180.0	179.0
status -j 550457c7c260573b8ecbf...	0.376
status -o operation	3.32	45.8	173.0	173.0	182.0
status -f {"a":{"$gt":-1}}	4.07
run --num-passes=1	15.5	66.9	201.0	203.0	203.0

signac 2.2.0, flow ce37f01

Command	operation	groupsof	groupsof_sort	groupby	groupby_sort
import	0.000208	0.000206	0.000206	0.000206	0.000202
instantiate	0.00584	0.44	1.17	0.729	0.82
status	2.59	1.57	2.87	0.868	0.967
status -j 550457c7c260573b8ecbf...	0.294
status -o operation	2.43	0.655	1.2	0.912	0.959
status -f {"a":{"$gt":-1}}	3.23
run --num-passes=1	14.5	1.94	2.63	1.11	1.15

All benchmarks run after signac update-cache.

Checklist:

• I am familiar with the Contributing Guidelines.
• I agree with the terms of the Contributor Agreement.
• My name is on the list of contributors.
• The changes introduced by this pull request are covered by existing or newly introduced tests.
• The package documentation and framework documentation in signac-docs are up to date with these changes.
• I have updated the changelog and added any related issue and pull request numbers for future reference.

[cbkerr]

Thanks!