Rust: Always force high access path precision

[hvitved]

Stages stats for AccessAfterLifetime.ql on rust before and after:

#	n	stage	nodes	fields	conscand	states	tuples	calledges	tfnodes	tftuples
1	10	1 Fwd	3,778,294	38,306	-1	1	4,503,242	-1	-1	-1
2	15	1 Rev	1,425,221	16,549	-1	1	1,667,531	1,470,786	-1	-1
3	20	2 Fwd	720,349	4,111	5,796	1	1,719,512	374,218	0	0
4	25	2 Rev	415,990	1,525	2,160	1	598,321	229,086	0	0
5	30	3 Fwd	158,978	1,008	17,292	1	3,800,765	208,204	19	167
6	35	3 Rev	141,824	843	9,830	1	937,772	203,758	9	20
7	40	4 Fwd	138,816	717	93,422	1	118,836,935	202,876	9	26
8	45	4 Rev	135,363	705	80,252	1	38,270,989	202,123	9	20
9	50	5 Fwd	127,950	566	22,131	1	27,011,587	200,142	9	26
10	55	5 Rev	109,938	458	7,423	1	1,584,901	192,141	9	17
11	60	6 Fwd	107,568	440	4,875	1	10,699,981	191,784	7	9
12	65	6 Rev	107,409	439	4,786	1	1,186,926	191,714	5	5

#	n	stage	nodes	fields	conscand	states	tuples	calledges	tfnodes	tftuples
1	10	1 Fwd	3,778,294	38,306	-1	1	4,503,242	-1	-1	-1
2	15	1 Rev	1,425,221	16,549	-1	1	1,667,531	1,470,786	-1	-1
3	20	2 Fwd	720,349	4,111	5,796	1	1,719,512	374,218	0	0
4	25	2 Rev	415,990	1,525	2,160	1	598,321	229,086	0	0
5	30	3 Fwd	158,978	1,008	17,292	1	3,800,765	208,204	19	167
6	35	3 Rev	141,824	843	9,830	1	937,772	203,758	9	20
7	40	4 Fwd	138,816	717	93,422	1	118,836,935	202,876	9	26
8	45	4 Rev	135,363	705	80,252	1	38,270,989	202,123	9	20
9	50	5 Fwd	114,126	513	3,200	1	681,843	192,053	7	24
10	55	5 Rev	91,447	392	1,076	1	122,901	182,333	7	7
11	60	6 Fwd	91,192	385	898	1	133,660	182,259	5	5
12	65	6 Rev	91,085	385	895	1	107,885	182,238	5	5

DCA confirms the speedup on rust.

[Copilot]

Pull request overview

Forces high-precision access-path tracking for Rust dataflow, disabling adaptive approximation to improve query performance.

Changes:

• Makes forceHighPrecision hold for all Rust dataflow content.
• Avoids costly adaptive access-path approximation.

Show a summary per file

File	Description
rust/ql/lib/codeql/rust/dataflow/internal/DataFlowImpl.qll	Enables high precision for every Rust access path.

Review details

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

• Files reviewed: 1/1 changed files
• Comments generated: 0
• Review effort level: Balanced