Allow passing int for uint spec when int precision > uint precision

[rly]

Fix #294.

Previously, when the ObjectMapper receives an int64 value for a uint32 spec, HDMF would raise an error. An int64 for a uint64, however, worked. The error was only raised if the given int value's precision was higher than the spec uint's precision.

With this change:

• if an int64/float64 value is given for a uint8/uint16/uint32/uint64 spec, the value will be converted to uint64
• if an int32/float32 value is given for a uint8/uint16/uint32 spec, the value will be converted to uint32

A warning will also be raised when the given base type differs from the specified type, such that a conversion occurs, e.g., a float32 value is given for an int32 spec. This was previously converted invisibly.

[codecov]

Codecov Report

Merging #298 into dev will increase coverage by 0.06%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##              dev     #298      +/-   ##
==========================================
+ Coverage   72.59%   72.66%   +0.06%     
==========================================
  Files          33       33              
  Lines        6390     6395       +5     
  Branches     1396     1398       +2     
==========================================
+ Hits         4639     4647       +8     
+ Misses       1322     1319       -3     
  Partials      429      429

Impacted Files	Coverage Δ
src/hdmf/build/objectmapper.py	65.89% <100%> (+0.68%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d432709...2676646. Read the comment docs.

[rly]

@oruebel mentioned that an int64 cannot be held within a uint32. int64 supports -2^63 to 2^63-1. uint32 supports 0 to 2^32-1.

@oruebel also suggested that a warning be raised if HDMF converts the given value to a different base type.

The PR has been updated to reflect these changes.