Bump rand to 0.9 and associated crates

[jakelishman]

Summary

This is a breaking API change, but our exposure is mostly just to some minor renames. Unfortunately the whole ecosystem hasn't migrated yet, so we're still compiling in multiple versions of rand, but with any luck, bumping hashbrown and dropping faer will help.

Details and comments

This obsoletes the three separate PRs that dependabot is trying to make (and keeps trying to jam up CI with):

Close #13745
Close #13746
Close #13747

[qiskit-bot]

One or more of the following people are relevant to this code:

• @Qiskit/terra-core

[jakelishman]

Oh ugh, looks like some of the seeding behaviour has changed as well.

[jakelishman]

Scanning through the changes to the pRNG we use between 0.8.5 and 0.9.0, and to the implementation of the standard uniform distribution, it doesn't look like we should have seen randomisation changes, which makes me a bit nervous. I wasn't expecting to see any - the mathematics of what makes up the pcg64_fast pRNG are fixed, and the internal seed mixing of those hasn't changed.

That said, running the tests locally several times shows that the randomness changes appear deterministic, so the chances are it's something I'm just missing in understanding. I'll push up the test changes to fix the random-seeded tests.

[coveralls]

Pull Request Test Coverage Report for Build 13330455581

Details

• 17 of 20 (85.0%) changed or added relevant lines in 8 files are covered.
• 11 unchanged lines in 4 files lost coverage.
• Overall coverage increased (+0.02%) to 88.245%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
crates/accelerate/src/circuit_library/iqp.rs	2	3	66.67%
crates/accelerate/src/synthesis/clifford/random_clifford.rs	5	6	83.33%
crates/accelerate/src/synthesis/linear/utils.rs	1	2	50.0%

Files with Coverage Reduction	New Missed Lines	%
crates/accelerate/src/unitary_synthesis.rs	1	93.29%
qiskit/providers/basic_provider/basic_simulator.py	1	96.99%
crates/qasm2/src/lex.rs	3	92.73%
crates/qasm2/src/parse.rs	6	97.15%

Totals
Change from base Build 13329011740:	0.02%
Covered Lines:	78709
Relevant Lines:	89194

---

💛 - Coveralls

[mtreinish]

Besides the sabre test file conflict this LGTM, I'll reapprove it once it's rebased. That being said I'm going to miss the from_entropy() function. I liked that name more.

[jakelishman]

Conflict should be resolved.