reversed "update" method in multi_cache to overwrite old values with new

[Trizalio]

If wrapped function can return dict with more keys, than required there is a case, in which new value will be replaced by old, example:

import asyncio
from typing import List, Dict

from aiocache import multi_cached


@multi_cached("ids", namespace="main")
async def foo(ids: List[int]) -> Dict[int, int]:
    biggest_value = max(ids)
    result = {id_: biggest_value for id_ in range(biggest_value + 1)}
    print(f'foo result: {result}')
    return result


async def amain():
    print(await foo(ids=[1]))
    print(await foo(ids=[0, 2]))
    print(await foo(ids=[0]))

if __name__ == "__main__":
    asyncio.run(amain())

will be printed:

foo result: {0: 1, 1: 1}
{0: 1, 1: 1}
foo result: {0: 2, 1: 2, 2: 2}
{0: 1, 1: 2, 2: 2}
{0: 1}

As we can see, on second call of foo value for 1 changed, but it was overwritten by cache value.

In my case i must return aggregated values based on values from two distinct bigdata storages. Based on requested keys i can predict future requests, so i'd like to preload values for predicted keys with keys passed to my funstion and save them all in aiocache

[argaen]

Thanks for this @Trizalio!

We would need to add tests in order to merge this

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.13%. Comparing base (a966f84) to head (9981377).
⚠️ Report is 362 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #489   +/-   ##
=======================================
  Coverage   99.13%   99.13%           
=======================================
  Files          13       13           
  Lines        1043     1044    +1     
  Branches      116      116           
=======================================
+ Hits         1034     1035    +1     
  Misses          9        9           

Files with missing lines	Coverage Δ
aiocache/decorators.py	100.00% <100.00%> (ø)

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update a966f84...9981377. Read the comment docs.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[Dreamsorcerer]

Looks reasonable, but as mentioned, please add a test (after merging master). Test probably should go in tests/ut/test_decorators.py