chore: bump fsrs version

[asukaminato0721]

Linked issue (required)

Refs #4080

Summary / motivation (required)

bump fsrs version

Steps to reproduce (required, use N/A if not applicable)

N/A

How to test (required)

Checklist (minimum)

• I ran ./ninja check or an equivalent relevant check locally.
• I added or updated tests when the change is non-trivial or behavior changed.

Details

Before / after behavior (optional)

Risk / compatibility / migration (optional)

UI evidence (required for visual changes; otherwise N/A)

Scope

• This PR is focused on one change (no unrelated edits).

[asukaminato0721]

cc @L-M-Sherlock

[david-allison]

What's the user-facing name of this release (6.3.0)?

Ref: https://github.com/ankidroid/Anki-Android/blob/27178821051b1a665fdea986807e4eb9428e8147/AnkiDroid/src/main/java/com/ichi2/anki/scheduling/Fsrs.kt#L42-L48

[asukaminato0721]

filled_fsrs_parameters

FSRS-6

[Expertium]

@asukaminato0721

1. I don't see release 6.5.0, only 6.4.1 nvm, it just got released
2. After Speed up FSRS inference, evaluation, and training without production Burn dependency open-spaced-repetition/fsrs-rs#411 and Speed up compute_parameters without SIMD open-spaced-repetition/fsrs-rs#419, we should use compute parameters with card IDs and compensate for the slight degradation in log loss by bumping the number of epochs from 5 to 8. The speedup will be better that way.
   I believe currently compute_parameters is 4-5x faster than before if we don't use card IDs and keep 5 epochs, and ~9x faster if we do use card IDs and increase epochs 5->8.

[L-M-Sherlock]

LGTM

[Expertium]

@L-M-Sherlock we can also include your suggestion to remove evaluation after each epoch aka best epoch selector and bump up the number of epochs by 1 to compensate. You said removing evaluation after every epoch makes it ~30% faster.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud