Optimizes GetAllAsync in HybridCacheClient by parallelizing local cache lookups and distributed cache population.
It also reduces network calls to the distributed cache by fetching all expirations in a single bulk operation.
This addresses performance issues related to excessive round trips and improves overall efficiency.

CoPilot session

Optimizes GetAllExpirationAsync in HybridCacheClient and InMemoryCacheClient by first checking the local cache and then retrieving any missed keys from the distributed cache to reduce distributed cache calls.

Adds null checks and optimizes the handling of empty key lists to improve efficiency and prevent errors.

…xpirations

Ensures that setting expirations for multiple cache keys
correctly publishes invalidation messages and handles empty
expiration dictionaries efficiently. This prevents unnecessary
message bus calls when no expirations are set, and guarantees
invalidation messages are sent with the correct keys after
setting the expirations on the distributed cache.

Optimizes the `GetAllAsync` method in `HybridCacheClient` to reduce latency and improve performance.

The previous implementation used parallel processing and thread-safe collections which introduced unnecessary overhead. This commit refactors the method to leverage bulk operations and minimize locking, resulting in more efficient cache retrieval. It now retrieves all local cache values at once, then fetches missing keys from the distributed cache, and finally populates the local cache with the distributed results. This reduces the number of network calls and improves overall throughput.

Updates the documentation to better explain when to use context7.

Specifically, it clarifies that context7 should be used when code generation,
setup/configuration steps, or library/API documentation is needed.

…ption types

Addresses inconsistencies in cache key validation by enforcing ArgumentException instead of ArgumentNullException for empty strings in RemoveIfEqualAsync.

Simplifies test logic by removing redundant ArgumentNullException checks for empty strings in GetListAsync and ListAddAsync.

Adjusts test parameters in RemoveByPrefixAsync to use a smaller value (9999) instead of (10000). This likely addresses edge case issues or performance considerations in the test environment.

Replaces multiple null checks with ArgumentNullException.ThrowIfNull
for improved code clarity and performance.

Also, replaces manual range checks with ArgumentOutOfRangeException.ThrowIfLessThan.
Adds Distinct to keys in RemoveAllAsync to ensure no duplicate keys.

Improves the ScopedCacheClient by adding argument validation
to prevent null or empty keys and values.

This ensures data integrity and prevents unexpected errors
when interacting with the cache.

Optimizes the GetAllAsync method in HybridCacheClient to improve performance by reducing unnecessary iterations and array conversions.
It now iterates through the initial collection of keys only once and avoids converting to an array unless necessary.
Additionally, it streamlines the process of identifying missed keys and fetching them from the distributed cache.

Improves cache performance analysis by adding a new benchmark to measure SetAllAsync/GetAllAsync throughput with a large number of keys.

This helps to evaluate bulk insert and retrieval performance and validates data correctness under high load.
Also, it updates existing benchmarks to report operations per second.

Updates RemoveAsync to return false if the entry was expired,
consistent with Redis behavior.

Updates the name of the ExistsAsync test method to more accurately reflect its purpose and the variety of scenarios it covers.

Renames the test methods to explicitly state that they handle large numbers correctly.

This change improves the readability and understanding of the test suite, making it clearer what specific scenarios are being tested.