Mid-capacity CPU cores are classified as Little on devices with an unusual core frequency distribution

Steps to reproduce:
1. Open the attached user's project "CoreDetection.zip"
2. Build and deploy to the affected devices
3. Observe the CPU classification in the table on the device

Actual result: Mid-capacity CPU cores are classified as Little
Expected result: Mid-capacity CPU cores are classified as Big

Reproducible with: 6000.0.82f1, 6000.3.23f1, 6000.5.10f1, 6000.6.0b10, 6000.7.0a5

Reproducible with these devices:
Google Pixel 11 Pro (via SauceLabs)
Google Pixel 11 (via SauceLabs)

Not reproducible on these devices:
VLNQA00692, Oppo Find X8 Pro (CPH2659), Android 16, CPU: MediaTek Dimensity 9400 (MT6991), GPU: Mali-G925-Immortalis MC12
VLNQA00661, Samsung Galaxy S26 (SM-S942B) - CPU: Exynos 2600, GPU: Xclipse 960, OS: 16
VLNQA00690, Xiaomi Xiaomi 15T Pro (2506BPN68G), Android 16, CPU: 1x Cortex-X925 @ 3.7GHz 3x Cortex-X4 @ 3.3GHz 4x Cortex-A720 @ 2.4GHz, GPU: Mali-G925-Immortalis MC12
VLNQA00580, Xiaomi Redmi Note 8 (M1908C3JGG), Android 11, CPU: Mediatek MT6769Z Helio G85 (12nm), GPU: Mali-G52
VLNQA00615, Samsung SM-S906B (SM-S906B), Android 13, CPU: Exynos 2200, GPU: Ltd.
N/A, Vivo - (V2183A), Android 12, CPU: NOT FOUND, GPU: Mali-G710 MC10
VLNQA00010, Samsung SM-G950U (SM-G950U), Android 9, CPU: Snapdragon 835 MSM8998, GPU: Adreno (TM) 540

Testing Environment: Windows 11
Not reproducible on: no other environment tested

Notes:
-Likely also affects Dimensity 9500 devices (X300 series, OPPO Find X9 series, Xiaomi 17T Pro, Galaxy Tab S12 Ultra)
-Suggestion: if only one core would be classified as big, also classify the next-highest-capacity cluster as big; adjust the threshold only when a single core would otherwise be classified as big