feat: get plan bulk with cache

[tomerqodo]

Benchmark PR from qodo-benchmark#445

[sentry]

Bug: The get_plan_bulk method returns None for invalid tenants, which is then cached as a JSON "null". This causes validation errors on subsequent cache reads.
_{Severity: HIGH}

Suggested Fix

Modify the get_plan_bulk method to not add entries for tenants with invalid plans to the results dictionary. Instead of results[tenant_id] = None, the code should simply continue or skip adding the key to the dictionary. This ensures the return value adheres to the dict[str, SubscriptionPlan] type hint and prevents None values from being cached.

Prompt for AI Agent

Review the code at the location below. A potential bug has been identified by an AI
agent.
Verify if this is a real issue. If it is, propose a fix; if not, explain why it's not
valid.

Location: api/services/billing_service.py#L292

Potential issue: In the `get_plan_bulk` method, when subscription plan data validation
fails for a given tenant, the code assigns `None` to the results for that tenant ID.
This `None` value is then passed to `get_plan_bulk_with_cache`, which attempts to cache
it. The caching logic serializes `None` into the JSON string `"null"` and stores it in
Redis. On subsequent cache reads for that tenant, the `"null"` string is deserialized
back to a `None` object, which then causes a validation failure when
`subscription_adapter.validate_python` is called, as `None` is not a valid
`SubscriptionPlan` object. This leads to repeated cache misses and unnecessary API calls
for invalid tenants.

_{Did we get this right? 👍 / 👎 to inform future reviews.}